Spread spectrum communication system and decentralized-control wireless network for implementing CDMA through application of a single spread spectrum code having different phase sequences

Field of Invention

The invention relates to a spread spectrum communication system for implementing Code Division Multiple Access (CDMA) through application of a single spread spectrum code having different phase sequences. The system is usable in a wireless network, especially for a decentralized-control wireless mobile communication network composed of mobile nodes.

Background of the Invention

Spread spectrum technique is used in two categories of wireless networks. Wireless networks of the first category, such as CDMA mobile communication systems, are centralized control networks, wherein each mobile station relies on a base station for dynamic allocation of spread spectrum codes (or information channels). The base station is necessary for the communication between two mobile stations (nodes) even though they are in proximity to each other. Wireless networks of the second category, such as autonomous wireless networks (wireless ad hoc networks) and wireless sensor networks, are decentralized controlled networks. Nodes in this category of networks do not require a base station or other means of supervisory control for communication therebetween and are directly communicable with one another. Additionally, when two nodes are unable to establish direct connectivity due to limited capacity or other reasons, communication between the two nodes is achieved through other nodes so that every node within the network is able to communicate with one another. As the nodes are mobile, the topological structure of this category of networks is dynamically changing. Therefore, this mode of communication between the nodes is not directly applicable to the communication networks of existing communication infrastructure. .

The use of spread spectrum technique in wireless networks has many advantages, especially in preventing signal attenuation, implementing CDMA and increasing processing capacity. Therefore, spread spectrum technique is utilized in many of today's wireless network for implementing communication systems. In direct sequence spread spectrum systems, what is essential is having suitable spread spectrum codes, which is the subject for developing the systems' key techniques. In CDMA mobile communication systems, mobile stations are synchronized so as to make use of cross-correlated spread spectrum codes for despreading and eliminating multiple access interference (MAI).

However, there are two problems concerning decentralized control wireless networks. The first problem is the difficulty in achieving synchronization for the entire wireless network. When synchronization of the entire wireless network is not achieved, and if Walsh codes are used, the cross-correlation between the Walsh codes becomes weakened and as a result, the MAI intensifies between communicating channels of the networks. If Gold codes are used, the cross- correlation between the Gold codes is improved, but because the number of communication channels is increased, the MAI inevitably intensifies between the communicating channels. In the case of Gold codes of length 127, the normalized cross-correlation value is 17/127, which is approximately 1/7. This value is not sufficiently large to provide significant improvements in network throughput. The m-sequence has a single autocorrelation peak value and desirable autocorrelation, but does not have a desirable cross correlation. The best cross-correlated sequence is a preferred-pair of m-sequences that has the same cross-correlation as the Gold codes. As seen from the case of the Gold codes of length 127, the largest cross correlation value is 17/127. It is thus clear that when synchronization is not achieved, using either the Walsh or Gold codes or different m-sequences in a decentralized control wireless network results in highly intensified MAI.

Additionally, it is very difficult to provide enough spread spectrum codes for generating a spread spectrum code set that has desirable cross correlation. As synchronization for the entire decentralized control wireless network is unlikely to be achieved, none of the Walsh codes, Gold codes and m-sequence is therefore

suitable for implementing CDMA. In selecting a spread spectrum code, it is earlier to achieve desirable autocorrelation with a single autocorrelation peak value, linear correlation and periodic correlation than generating a spread spectrum code set that has desirable cross-correlation. Baker code, m-sequence and specific spread spectrum codes with cross correlation is zero all have desirable autocorrelation. For instance, in m-sequence, which has desirable autocorrelation, the cross-correlation value between an m-sequence PN(t) and a shifted m-sequence PN(t-kτ _c ) is —UN when the value of the cyclic shift kτ _c is larger than the period of a time slot. This is necessary for the formation of code channels.

In existing decentralized control wireless networks that adopt the spread spectrum technique, such as ad hoc sensor networks, there is one type that employs a single spread spectrum channel on a physical layer and a medium access control layer that conforms to the 802.11 DCF protocol. This type of networks do not fully make use of the advantage of using spread spectrum code division for generating multiple channels and increasing processing capacity. Another type of decentralized control wireless networks makes use of the spread spectrum code division for generating multiple channels, but each node within the networks is allocated a fixed spread spectrum code. A first and second conventional multichannel spread spectrum communication systems for use in a decentralized control wireless network are shown in Figs. 1 and 2 respectively.

An access method adopted by the first and second conventional multichannel spread spectrum communication systems is described hereinafter. When a source node within the systems has data for transmission, an RTS packet is transmitted to a common channel C ₁ , where idle nodes on the common channel C ₁ uses spread spectrum codes Ci for despreading the RTS packet. A destination code on the common channel Cj receives the RTS packet and sends CTS packet to the source node. The CTS packet contains information such as source node identification address (or ID), destination node ID and channel code used for transmitting a data packet. After receiving the CTS packet, the source node then transmits the data packet through a selected channel Ci and uses spread spectrum codes Cj for modulating the data contained in the data packet. Immediately after the destination

node on the common channel has completed the transmission of the CTS packet, the destination node switches to the selected channel Cj as defined by the channel code for receiving the packet and applying the spread spectrum codes Ci for despreading the data. Once the destination node completes the reception of the data packet, an acknowledgement packet is transmitted on the same selected channel Cj by the destination node. Other nodes on the common channel that receives the RTS packet or CTS packet automatically avoid the selected channel Cj for a period that is determined by time lengths of the data packet and the CTS packet as well as the period of a protection time-slot. The access method is completed at this juncture. The first and second conventional multiple channel spread spectrum communication systems is capable of implementing dynamic code allocation (DCA).

The operational principal for implementing the spread spectrum communication system of Fig. 2 for use in a decentralized control wireless network is the same as that previously described. The difference between the spread spectrum communication systems of Figs. 1 and 2 is that the system of Fig. 2 has m number of matchers that correspond to C ₁ C _m of the spread spectrum codes for performing despreading. The system as illustrated in Fig. 2 has a larger and more complex hardware configuration.

The matcher of Fig. 2 is further illustrated in Fig. 3. The matcher completes the following function :

JV-I

R(n) = ∑a,x _l+n

(=0 where {a;} (0< i < N-I) is the local coefficient series, {xi} (0 < i < ∞) is the reception series and R(n) represents the calculated correlation value. When the local coefficient series {aj} obtains a value of {±1}, the reception series after A/D conversion becomes a quantized numerical series.

The two previously described spread spectrum communication systems have the following disadvantages when used in decentralized-control wireless networks:

(1) When synchronization of the entire wireless network is not achieved, obtaining desirable cross-correlation among the set of C ₁ C _m spread spectrum codes is practically impossible.

(2) Due to the mobile nature of the decentralized-control wireless networks, instances of collision on common or data channels are inevitable.

(3) Channels for transmitting data are selected by the source nodes, without reference to any channel state tables of corresponding destination nodes for accessing the state of each channel, and if the state of the channels that are selected by the source nodes is busy, this inevitably causes access failure.

(4) Unresolved problems related to hidden and exposed terminals.

(5) Not provision of real-time monitoring of the state of channels.

There is another conventional type of conventional communication system used in decentralized control wireless network. The system clock for all nodes in such a communication system is obtained from a satellite clock signal, which is received by a satellite receiver for receiving telecommunication codes transmitted from a satellite, and where clock signals are extracted from the wireless telecommunication codes to be used as the system clock for the nodes, (see 'SYN-MAC: A Distributed Medium Access Control Protocol for Synchronized Wireless Networks', www. cacs . louisiana. edu, Hongyi Wu, Anant Utgikar, and Nian-Feng Tzeng and 'A Simple Distributed PRMA for MANETs', IEEE Transactions on vehicular technology, Vol51, No2 MARCH 2002 P293-P305) see Fig. 4. The key feature of applying such a scheme in a communication system is that all nodes within the entire wireless network receive the same clock signal from the satellite and this ensures that signals are transmitted synchronously. However, because the nodes are randomly distributed mobile in nature, it can not guaranteed that signal channels are orthogonal to one another when signals have reached the destination nodes. This means that multiple access interference is not fully eliminated.

Accordingly, because each node of the decentralized-control wireless network is mobile and autonomous, the network topology changes dynamically and this results in the inability of the entire network to achieve synchronization. Therefore, directly

applying conventional CDMA technique in the decentralized control wireless network for implementing multiple access is not feasible. When the network is not synchronized, the degree MAI between channels is intensified. Therefore, in present decentralized control wireless networks, a single spread spectrum channel on a physical layer and a medium access control layer that conforms to the 802.11 DCF protocol are used. Due to decentralizing control, and the lack of knowledge in the state of channel usage of the entire network on the part of mobile nodes, the technique of multiple codes CDMA when used for implementing multiple access channels causes the DCA scheme to be impractical, which greatly limits the application of spread spectrum technique in wireless networks.

Summary of the Invention

A communication system for implementing code division multiple access in a wireless communication network is disclosed. The communication system comprises a first processing unit for receiving a clock-containing signal and for extracting clock signals from the clock-containing signal. The communication system further comprises a second processing unit for receiving and processing data signals, wherein when in use, the second processing unit receives the clock signals from the first processing unit for synchronizing the processing of the data signals, utilizes at least two phase sequences of a single spread spectrum code generated in the second processing unit for generating spread spectrum communication signals, and transmits the spread spectrum communication signals.

Brief Description of Drawings Embodiments of the invention are described in detail hereinafter with reference to the drawings, in which: . »

Fig. 1 is a schematic diagram of a conventional multichannel spread spectrum communication system that is used in wireless network;

Fig. 2 is a schematic diagram of another conventional multichannel spread spectrum communication system that is used in wireless network;

Fig. 3 is a schematic diagram of a matching unit according to an embodiment of the invention;

Fig. 4 is a schematic diagram of the embodiment of the invention that utilizes a satellite-received clock extractor;

Fig. 5 is a schematic diagram of a decentralized control wireless network according to the embodiment of the invention;

Fig. 6 is a schematic diagram of a spread spectrum communication system that utilizes different phase sequences of a spread spectrum code for implementing code division;

Fig. 7 is a schematic diagram of spread spectrum code generator;

Fig. 8 is another structure diagram of the matching unit;

Fig. 9 is a schematic diagram of a control time-slot generation unit; and

Fig. 10 is graph showing control time slot signals.

Detailed Description of the Preferred Embodiment

The purpose of the invention is to provide a spread spectrum communication system for use in a wireless network. The spread spectrum communication system utilizes different phase sequences of a single spread spectrum code and allows mobile nodes (nodes) under the condition of decentralizing control to realize Code Division Multiple Access (multiple channels) and establish linkage between any two nodes. Multiple pairs of nodes (each pair of nodes is a pair of transmission and reception nodes) utilize the different phase sequences of a single spread spectrum code for performing spread spectrum on data signals. The nodes use data clock provided by a satellite so that within the coverage area of each node, each bit of transmitted signals is essentially synchronized. As long as the path delay between source and destination nodes within the coverage area thereof is shorter than the time-delay of adjacent phase

sequences, co-channel interference at the destination node is averted. This allows identical frequency bands to be used simultaneously for wireless communication and as a result, improves the processing capacity of the wireless network and reduces multiple access interference therein.

Another purpose of the invention is to provide a decentralized control wireless network that uses the spread spectrum communication system. Because each of the different channels utilizes a different phase sequences of a spread spectrum code, every node within the wireless network is capable of autonomous monitoring of the state of channel usage within the coverage area. This allows self-implementation of dynamic distribution of channels (self-selection of idling spread spectrum codes) to be achieved so that collision therebetween is avoided.

In order to achieve the above purposes of the invention, the concept of the invention is described in greater detail hereinafter.

The objective of the invention is to reduce multiple access interference in decentralized control wireless network through the implementation of a dynamic code allocation (DCA) mechanism. Essentially, all nodes receive wireless signals from a satellite, wherein a clock is extracted from the wireless signals for providing data clocking for the nodes. A spread spectrum code set composed of different phase sequences of a single spread spectrum code is used. With the use of a centralized data clock, synchronization of the transmission of spread spectrum signals between nodes within the coverage area is established. As long as the path delay between source and destination nodes within the coverage area is shorter than the time-delay of an adjacent phase sequences, any co-channel interference is averted.

Theoretically, the path difference between satellite signals that has traveled a distance r ² r is d » -— , where R is the distance between the satellites and _. Earth. Assuming that

the radius of the coverage area of a node within a decentralized control wireless network is 300m and that the satellite is a low orbit satellite (R=300Km), then d ~

0.15m and the path delay between two satellite signals is τ ~ d/C = 0.5ns. The time delay in transmitting signals from any two nodes, A and C, within the coverage area of a destination node to the destination node is not more than Ins. If the data transfer rate is IMbps, utilizing m-sequence spread spectrum codes of length 63 gives T _c K 16ns. This suggests that the effects of path delay within a coverage area on the delay of a spread spectrum signal is small, which is 1/16 of a Chip time slot. It can be assumed that within the decentralized-control wireless network, the signal transmission between two nodes within the coverage area of any node is synchronized. As the time delay is not significant, the cross correlation value is small (such as the cross correlation value of the m-sequence having a phase shift that is greater than the Chip time-slot) and that multiple access interference (MAI) between the two nodes is negligible. Different nodes are able to receive the required data signals transmitted over a channel corresponding to the phase sequence. Critically, as each node uses a different phase sequences of a spread spectrum code for performing spread spectrum, each of the nodes are able to assess the state of all channels, thereby realizing decentralized control and the mechanism of dynamic code allocation (DCA).

The invention involves the use of spread spectrum techniques in wireless networks, especially decentralized control wireless networks. The networks consist of a plurality of mobile nodes, where all nodes within the coverage area of a node is able to engage in half duplex communication (time division duplex TTD method) with that node. Other nodes that are outside the coverage area are unable to communication directly with that node, unless communication is established indirectly through an intermediate node.

Fig. 5 shows a network structural chart of a spread spectrum communication system according to the invention. Each node 100 in the network structural chart has identical hardware and software configurations for used in the spread spectrum communication system.

In accordance to the previously described concept of the invention, a technical scheme of the invention is described as follows:

A type of spread spectrum communication system that uses phase sequences of a single spread spectrum code for implementing code division multiple access, including: a matching unit for receiving wireless signals, the matching unit transmits multipath related data; a related multipath selection unit under the control of a processing control unit, the related multipath selection unit transmits related data during a specified time slot to a demodulation unit; a demodulation unit for demodulating baseband information carried by the related data during a specified time-slot; a duplexer, the duplexer being under the control of the processing control unit for completing transmission and reception; and a spread spectrum modulation unit. The spread spectrum modulation unit expands a low speed narrowband signal into a broadband signal. Under the control of the processing control unit, the spread spectrum modulation unit cooperates with a code multipath selection unit. When the system transmits control packets, the spread spectrum modulation unit selects common channels that correspond to the phase sequence output of a spread spectrum code. When transmitting data packets, the spread spectrum modulation unit transceives the phase sequence output that is arranged between a transmitter and a receiver; the processing control unit, which controls the related multipath selection unit, the demodulation unit, the duplexer and the spread spectrum modulation unit. The processing control unit further comprises: a clock extracting unit for receiving wireless signals from a satellite and extracting clock signals from the wireless signal for use as a data clock for the system; a clock synchronization unit for generating a spread spectrum code clock, the spread spectrum code having a length of N, the relationship between the period Tj of the data clock and the period T _c of a spread spectrum chip (in short, chip) is T _d ⁼ NT _C and this ensures the synchronization between data and spread spectrum codes; a spread spectrum code set generator for generating Baker codes, m-sequences or other code sequences that have desirable self-correlation. The spread spectrum codes set generator generates code sets composed of the spread spectrum code and the phase sequence of the spread spectrum codes.

A time-sequence generating unit for generating time sequence signals that corresponds to a channel (spread spectrum code phase sequence).

The processing control unit also includes a CPU processor. The CPU processor fetches a specific correlation value in a specific time slot. The magnitude of the correlation value represents the state of traffic of a channel. The CPU processor compares the correlation value with upper and lower threshold values. If the correlation value exceeds the upper threshold value, the channel is busy. The channel is idle if the correlation value is less than the lower threshold value. This comparison is stored in a channel state table and is being constantly undated. At the same time, the CPU processor stores the protocol for Medium Access Control (MAC). When under the control of the processing control unit, the CPU processor causes an idling demodulator to specifically demodulate information on a common channel. After completing the negotiation over a common channel, a node subsequently transmits data information over a channel as specified during the negotiation. During transmission of control and data packet, the node selects, respectively, a first spread spectrum code phase sequence corresponding to a common channel and a second spread spectrum code phase sequence as specified during the negotiation. The processing control unit receives data from the demodulation unit and transfers the data to an upper program for processing. Additionally, the processing control unit controls the completion of transceiving data by the duplexer and both the selections of spread spectrum codes and multipath related outputs.

The spread spectrum communication system utilizes a spread spectrum set that is composed of different phase sequences of a spread spectrum code, such as spread spectrum codes PN(t), PN(t-τ ₀ ) PN(t-k τ ₀ ), where N is the code length and that the extraction of k should satisfy (k+1) τ ₀ <NT _C .

In the spread spectrum communication system, the matching unit transmits respective matching correlation values RO, Rl, R2 Rk of the multipath and spread spectrum codes PN(t), PN(t-τo) PN(t-k τ ₀ ).

The clocking of the spread spectrum chip (Chip) and the clocking of the data are synchronized. The periods of the data and Chip clocks are related as follows: T _d = NT _c , where N is the length of the spread spectrum code.

As a single spread spectrum code is used in the spread spectrum communication system, different phase sequences of the spread spectrum code represent different channels. This allows the matching unit 107 to receive and transmit spread spectrum code signals and multipath correlation values respectively, and detecting the state of traffic of different channels by fetching the correlation peak values of the multipath during different time slots. The channel is busy if the correlation value exceeds an upper threshold value while the channel is idle if the correlation value is less than a lower threshold value.

The decentralized-control wireless network utilizing spread spectrum technique is composed of a plurality of nodes, wherein each node is the spread spectrum communication system as previously described.

In the decentralized-control wireless network, every node receives the same wireless signal from a satellite and extracts a clock signal from the wireless signal for use as a data clock signal for the system.

The decentralized-control wireless network, wherein the relationship between the radius of a coverage area of each node therein and the time delay between two adjacent phase sequences of the spread spectrum code set is as follows: r < C τ ₀ , where C is the speed of light and τ ₀ is the time delay between two adjacent phase sequences. Within the coverage area of radius r, each node is capable of accurately receiving information transmitted from the decentralized-control wireless network. Beyond the coverage area, information transmitted from the decentralized-control wireless network is not accurately received.

Within the decentralized-control wireless network, as each node is able to monitor the state of usage of each of the spread spectrum channels, the decentralized-control wireless network is capable of performing dynamic distribution of signals without the

need of a base station. The distribution mechanism is described hereinafter. A source node transmits a request packet over a common channel for requesting the transmission of a data packet, wherein the request packet contains identification address (hereinafter ID) of the source node, the ID of a reception node and the channel state table of the source node. Once the destination node receives the request packet over the common channel, the destination node compares the channel state table of the source node with its own, and randomly selects an idle channel. Thereafter, the destination node transmits a response packet over the common channel, wherein the response packet contains the ID of the source node, the ID of the destination node and the code of the selected channel. After the source node receives the response packet, the source node transmits an acknowledgement packet over the common channel, wherein the acknowledgement packet contains the ID of the source node, the ID of the reception node and the channel code. The source node then transmits the data packet over an appropriate channel. After the destination node has received the acknowledgement packet over the common channel, the destination node then switches to the appropriate channel to receive the data packet. After completing the reception of the data packet, the destination node then transmits the acknowledgement packet over the appropriate channel.

As the data clocks of all nodes within the decentralized wireless network are essentially synchronized, the nodes are able to use the same spread spectrum code set, which is composed of different phase sequences that applies the m-sequence or other spread spectrum codes having a single cross correlation peak value. The implementation method involves the followings: a spread spectrum code PN(t) and a spread spectrum code set which consist of different phase sequences: PN(t), PN(t-τo),

PN(t-2τ ₀ ) ) PN(Uc τ ₀ ), wherein k τ ₀ < NT _C . τ ₀ is a multiple of T _c and τ ₀ = J T ₀ .

The number of codes within the spread spectrum code set is I = [N/J]. The selection of J involves the coverage area of a node having a radius r < C τ ₀ . If the coverage area is large, J is appropriately increased so as to ensure that the spread spectrum codes that are used by any two nodes are not affected by path delays and that the two nodes do not receive the same spread spectrum codes, thus avoiding co-channel interference.

The invention allows a node to perform real-time monitoring of all channels. As the entire network uses a single spread spectrum code and different phase sequences for representing different channels, the use of a matching unit therefore enables transmission of multipath correlation data. The presence of a cross correlation peak value within a specified time-slot signifies that a respective channel (phase sequence) is busy while the lack of any cross correlation peak values that is less than a specified threshold value suggests that the respective channel is idle. At the same time, each node monitors the maxima of the correlation peak value such that when the correlation peak value is larger than a threshold value, the separation between two nodes is small and that the two nodes are entering into each other's area of multiple access interference. When this happens, the two nodes should avoid simultaneous transmission or reception.

Every node has a channel state table, which stores the state of the traffic of all channels. The channel state table facilitates the selection of the idle channel by the nodes for multiple accessing.

Each node further monitors its separation from neighboring nodes and includes a separation table that stores the correlation peak values of signals received from the neighboring nodes. When the correlation peak value is larger than a threshold value, the separation between two nodes is small and this may indicate that the two nodes have entered into each other's area of MAI. When this happens, multiple access interference is avoided if one of the nodes receives signals (communicating) when the other retreat (retract).

The method of performing multiple access of each node is as follows: Each node performs dynamic code allocation (DCA). When a node has data to transmit, that node first completes CSMA/CA access over a common channel (for example using PN(t) for performing spread spectrum). If the channel is busy, no further action is taken over a period of one time slot. Thereafter, the channel is assessed of its availability. If the channel is still busy, no further action is taken until the next period is over. Subsequently, if the channel is available, a request packet is then transmitted. Nodes that are idling wait on the common channel. The request packet contains an ID

of a source node, an ID of a destination node and a source node state signal.

After the destination node has received the request packet from the source node, the destination node assesses the source node state signal and, according to information on the state of the channel of the source node, selects a channel signal that indicates the availability of both the source and destination nodes and includes the channel signal in a response packet to be transmitted from the destination node. After the source node has received the response packet, the source node subsequently transmits an acknowledgement packet (ACK), which includes a signal that provides information on selectable transmission channels.

If the destination node is not within the coverage area of a source node, a routing protocol is then used to perform routing. The source node and neighboring nodes of the destination nodes wait at an idle channel and monitor information on the RTS packet, CTS packet and the acknowledgement packet so as to avoid the same channel that is used by the source and destination nodes.

After the source nodes has transmitted the acknowledgement packet ACK ₅ the source node switches to a selected channel for transmitting a data packet while the destination node switches to the selected channel to receive the data packet. Once the data packet is received, the destination node transmits the acknowledgement packet over the selected channel before releasing that channel.

As compared to conventional techniques, the present invention has the following unique features and advantages: (1) The present invention uses a single spread spectrum code, where it is easier to obtain a single spread spectrum code having desirable autocorrelation than a code set that has desirable cross-correlation. _. The invention is also easier to implement and is less susceptible to multiple access interference. (2) Ease of implementing multichannel access and a larger processing capacity. (3) Nodes are capable of performing real-time monitoring of the state of channels.

(4) As different channels use different phase sequences of a spread spectrum code, each node is capable of self-implementating dynamic distribution of channels

(self-selection of idling spread spectrum codes) such that collision therebetween is avoided.

(6) As nodes are capable of performing real-time monitoring of the state of channels, problems related to the hidden and exposed terminals are resolved.

Practical implementation method

A preferred embodiment of the present invention is described hereinafter with reference to the drawings:

Fig. 5 is a schematic diagram of a decentralized control wireless network that performs different phase sequence of a single spread spectrum code for implementing code division multiple access (CDMA). Nodes A and B, C and D are two pairs of communicating nodes. Nodes H and E are neighboring nodes of node A, but are idling and therefore not transmitting any data. Except for nodes H, B and E, the other nodes are outside the coverage area of node A and are unable to directly receive data from node A, unless the data is retransmitted through other nodes. For example, data may be transmitted to node O through node B. In typical wireless networks, node E is unable to communicate with node F. In the present invention, nodes E and F are able to communicate with each other at different communication channels.

Fig. 6 shows a schematic diagram of a spread spectrum communication system, which is represented by a node. The system comprises a satellite-received clock extractor, a clock synchronization unit, a PN code generator, a control time-sequence unit, a spread spectrum modulation unit, a shaping circuit, a high frequency front-end circuit, a duplexer, a processing control unit, a matching unit and a channel monitoring unit.

The spread spectrum communication system has a practical structure as follows:

(1) The satellite-received clock extractor 101 receives wireless signals through a satellite antenna 112 and extracts clock signals from the wireless signals for use as a data clock. The data clock is sent to the satellite-received clock extractor 102, the time sequence control generation unit 111, the PN code generator 103 and a processing control unit 110.

(2) The satellite-received clock extractor 102 comprises a phase detector 102-1, a low pass filter 102-2, a voltage control oscillator 102-3 and a 1/N frequency divider 102-4, as shown in Fig. 6. The data clock is linked to an input terminal of the phase detector 102-1, which sends the data clock to an input terminal of the low pass filter 102-2 and further to the voltage control oscillator 102-3. The voltage control oscillator 102-3 outputs a Chip clock signal, which is sent to an input terminal of the PN code generator 103, the control time-sequence generation unit 111 and the processing control unit 110 respectively. At the same time, the voltage control oscillator 102-3 sends the

Chip clock signal via another output terminal thereof to an input terminal of the 1/N frequency divider 102-4, which then sends the Chip clock signal to an input terminal of the phase detector 102-1. In this manner, the phase detector 102-1, the low pass filter 102-2, the voltage, control oscillator 102-3 and the 1/N frequency divider 102-4 form a phase-locked loop. Through the phase- locked loop, the period of the input Chip clock pulse is 1/N that of the period of the input clock pulse.

(3) The PN code generator 103 generates a spread spectrum sequence code set {PN(t), PN(T-T ₀ ) PN(t-k τ ₀ )}, where K+l outputs are sent to an input terminal of a code multipath selection unit 104. The PN code generator 103 is connected to the processing control unit 110 via an input control bus. Under the control of the processing control unit 110, the code multipath selection unit 104 selects a specific phase sequence for transmission.

(4) An input terminal of a spread spectrum modulation unit 105 receives data outputs from the processing control unit 110 while another input terminal of

. the spread spectrum modulation unit 105 receives outputs from the code multipath selection unit 104. After the spread spectrum modulation unit 105 has completed the spread spectrum modulation of data signals, the spread spectrum modulation unit 105 outputs pulse signals that have a pulse rate that is N times faster than the pulse rate of the data signal.

(5) A duplexer 106 is connected to an antenna 113. The duplexer 106 has an input terminal for transmitting signals and an output terminal for receiving signals and an input control terminal that is connected to an output control

terminal of the processing control unit 110. Under the control of the processing control unit 110, the duplexer 106 completes the TDD method of transceiving.

(6) A matching unit 107 has an input terminal for receiving sequence signals of sampled analog signals. The matching unit 107 has output terminals for matching (k+1) paths with related data sequences R ₀ R _k , wherein the (k+1) paths receives the related (k+1) inputs from a related multipath selection unit 108 and the processing control unit 110.

(7) The related multipath selection unit 108 has (k+1) paths, which is connected to respective (k+1) paths of the processing control unit 110. The related multipath selection unit 108 has an input control terminal that is connected to the output control terminal of the processing control unit 110. The related multipath selection unit 108 further has (k+1) control terminals that are connected to respective (k+1) time slot control terminals To T _k of a control time slot generation unit. Under the control of the processing control unit 110 and related control time slots, the related multipath path selection unit 108 selects related outputs during specified time slots

(8) A demodulation unit 109 is connected to the related multipath path selection unit 108 and sends demodulated outputs to the data input terminal of the processing control unit 110.

(9) A control time sequence generation unit 111 is connected to the satellite- received clock extraction unit 101 for receiving the data clock therefrom. The control time sequence generation unit 111 is also connected to the Chip clock synchronization unit 102 for receiving the chip clock therefrom. The control time sequence generation unit 111 sends (k+1) control time-slot signals

To T _k to each of the related multipath selection unit 108 and processing control unit 110.

(10) Under the control of the processing control unit 110, the code multipath selection unit 104 selects a sequence from the sequence set (PN(t), PN(t-τo) PN(t-k τ ₀ )} generated by the PN code generator 103 and sends the sequence to the spread spectrum modulation unit 105 for modulation. The related multipath selection unit 108 selects a path Rj from the (k+1) outputs R ₀ R _k of the matching unit 107 and sends data within time slot T; to the

demodulation unit 109, which subsequently demodulates the data and sends the demodulated data to the processing control unit 110. At the same time, within a period Ta, the processing control unit 110 reads data R ₀ within time slot T ₀ , data R ₁ within time slot T ₁ and until data R _k within time slot Tic, and determines if the data within a time-slot exceeds a specified threshold value. Exceeding the threshold value suggests that a relevant channel is busy, otherwise the relevant channel is idling. Such information is stored in a channel state table. For example, when data Rj within time slot Tj exceeds the threshold value, channel i is busy, and otherwise channel i is idling.

Fig. 7 shows the schematic diagram of the PN code generator 103 of Fig. 6. The PN code generator 103 is implemented as follows:

The PN code generator 103 comprises k number of J-stage shift registers that are serially coupled to an output terminal of a conventional spread spectrum code generator, wherein J(k+1) < N. The time delay shift of a J-stage shift register is τo — JT _C and through the time delay shifting of each J stage shift register, the output from each respective level is PN(t), PN(t-τ ₀ ) PN(t-k τ ₀ ). The J-stage shift register as shown in Fig. 7 provides a basic unit. Each J stage shift register is composed of J number of D flip-flops, wherein a first D flip-flop 103-2 has an input terminal that receives output PN(t) from the spread spectrum code generator and an output terminal that is connected to an input terminal of a next stage D flip-flop 103-3. Accordingly, the output at the J ^λ D flip-flop 103-4 is a phase sequence PN(t-k τ ₀ ) of the spread spectrum code PN(t). At the same time, the D flip-flop 103-4 outputs the phase sequence PN(t-k τ ₀ ) to an input terminal of the next J-stage shift register and so on.

The matching unit 107 is implemented as follows:

Fig. 8 shows an input series {XJ} of the matching unit 107, wherein the coefficient sequence is {an-i, aN _-2 , ^•■■ ar-au a ₀ , aN-h aN-2, "'SN- _J K) - The input series has the following physical meaning: the input series performs sample quantization on received signals to obtain a binary bit sequence. The coefficient sequence has the following physical meaning: the coefficient sequence indicates the fixed coefficient that is multiplied to each matching filter of the matching unit 107. The coefficient

sequence corresponds to a one-bit binary system, which has a sequence consisting of 0 and 1. When the fixed coefficient is 0, this means that a corresponding multiplier can be omitted and that the method of realizing the matching unit 107 includes the following steps: (1) Under the drive of clock signals and starting from xo, the input series is sent to a time delay unit 107-1.

(2) Through the time delay units of each stage, respective signals are simultaneously sent to all specified coefficient multipliers via an input terminal thereof. (3) Each coefficient multiplier has another input terminal for receiving each specified coefficient corresponding to the matching unit 107. This means that au-i is sent to an input terminal of a first stage multiplier 107-6, a^ is sent to an input terminal of a second stage multiplier 107-7, and accordingly, a ₀ is sent to an input terminal of an N stage multiplier 107-8. Coefficient a _N -i is sent to an input terminal of the N+l stage multiplier, while coefficient aκ _-2 is sent to an input terminal of an N+2 stage multiplier. Accordingly, an N+JK stage multiplier has an input terminal for receiving coefficient a _N -jκ- It can be seen that the matching unit 107 has N+JK stages of time delay units.

(4) Through the multiplication operation of each 'multiplier, the respective results, y _N -i, V _N - ₂ , -yo, z _N-1 , z _N-2 , -Z _N . _JK , of the multiplication operation are simultaneously sent to a corresponding input terminal of a summing network 107-11.

(5) The summing network has k number of output terminals, which are respectively:

N

*o = ∑JV-«

;=1

N-jJ JJ Rj = ∑ I=I y»→ ⁺ ∑ BI=I ^z »-

l ≤j ≤ k

where R ₀ corresponds to spread spectrum code matching outputs of PN(t) = {a ₀ , a;— aN-i}, R ₁ corresponds to spread spectrum code matching outputs of PN(t-To) = PN(t- JT ₀ ) = { aN-j, S _N -J+ _I — aκ-i, ao, a} ^••■ until position J is circulated. The other output terminals are on the analogy of this.

The control time sequence generation unit 111 is implemented as follows: The control time-slot generation unit 111 as shown in Fig. 9 generates Chip clock pulses T _c through an output terminal thereof, and receives data clock pulses Td through an input terminal thereof.

The rising edge of the data clock T _d resets an N-bit counter 111-1. The N-bit counter 111-1 counts the number of incoming Chip clocks T _c and sends that number through a carry terminal C. The control time sequence generation unit is composed of J stage shift registers and OR circuits, as shown in Fig. 9. Basic units of the control time sequence generation unit are described as follows: The first basic unit has an input terminal that is connected to the carry terminal of the N-bit counter 111-1. The first basic unit is a D flip-flop 112-2 having an input terminal that is connected to the carry terminal of the N-bit counter 111-1. The first D flip-flop 112-2 has an output terminal that is connected to an input terminal of a second D flip-flop 112-3, and so on. The J ^th D flip-flop 111-4 has an input terminal that is connected to an output terminal of a J-I stage D flip-flop, which is connected to an input terminal of the next basic unit, and so on. The clock input terminals CP of all shift registers are connected to a Chip clock pulse T ₀ . Control time slots TS ₀ are generated as follows: The J number of input terminals of the OR circuit 111-5 are respectively connected to the J number of input terminals T ₁ , T ₂ Tj of the first J stage D flip-flop. Other control time slots

TSi TSK are on the analogy of this. The corresponding relationship between time

slot signals generated by the matching unit of Fig. 8, the Chip clock pulse T ₀ and the data time slot T^ is shown in Fig. 9.