MODULATION OF SIGNALS

This invention relates to modulation of signals. In particular, the invention relates to an OFDM modulation method, to a method of controlling the peak-to-average power ratio of an OFDM modulated signal, to a method of demodulating a modulated OFDM signal, to an OFDM modulator, to an OFDM demodulator and to an OFDM communication arrangement.

BACKGROUND OF THE INVENTION The inventors are aware of an OFDM signal schemes. However, a major disadvantage of an OFDM signal, is it's high peak-to-average power ratio (PAPR). Various methods have been suggested to reduce the PAPR, which includes clipping, coding, non-linear companding transforms, partial transmitted sequence, selective mapping, active constellation extension, tone reservation and constant envelope phase modulation. The inventors sought a method which requires low implementation complexity, which does not require any additional bandwidth expansion and which does not require side information to reconstruct the original message signal. Furthermore, the method should not lead to a severe bit error rate degradation as the number of carriers increase. The present invention aims to address these requirements and is descriptively referred to as offset modulation of an orthogonal frequency division multiplexing (OM-OFDM) signal.

SUMMARY OF THE INVENTION

According to one aspect of the invention, there is provided an OFDM modulation method, which includes

phase modulating the real and imaginary components of an OFDM signal independently. The method may include the prior step of independently scaling the real and imaginary components of the OFDM signal.

The scaling of the real and imaginary components of the OFDM signal may be given by:

Φ ₂ (ί) =— ί-— ^ where ¾ and 3 respectively refer to the real and imaginary components of the OFDM message signal, ς refers to a constant division term and where Oi(t) and Φ2(ϊ) represent the scaled real and imaginary OFDM components. These Oi(t) and Φ2Ο) are interchangeable.

The method may include the prior step of obtaining an inverse Fourier transform (IFT) of the OFDM message signal. The IFT may be an N - point inverse Fourier transform given by:

where T* is the symbol duration and Xk represents the complex signal written The method may include the step of adding an offset term to the scaled real component of the OFDM signal.

The method may include the further step of subtracting the phase modulated OFDM scaled imaginary component from the phase modulated OFDM scaled real component.

The phase modulated addition of an offset term to the scaled real portion of the OFDM signal and the subtraction of the phase modulated OFDM scaled imaginary component can be referred to as offset modulation of an orthogonal frequency division multiplexing (OM-OFDM) signal and given by: cos(2nfct + Φι(ή + ψ _0ί ) - cos(27 ct + Φι(ί)) where f is the carrier frequency, Φι(ί) and Φ2(ϊ) represent the scaled real and imaginary OFDM components and ψ _0ί represent the offset term.

The offset term ^x ¥ _os may be chosen such that ^x ¥ _os is sufficiently large and Φι(ί) and Φ2Ο) are sufficiently small. Typically, may be approximately twenty two times larger than Φι(ί) and Φ2Ο) . The combination of ^x ¥ _os and ς terms ensure that the receiver can successfully detect the original signal.

The method may include the further step of adding or subtracting a dominant frequency component to the modulated OFDM signal.

The dominant frequency component may be given by:

1 where γ is a dominant frequency component control factor, β is the adapted phase deviation of the real and imaginary OFDM signal and Jo( ?) is a Bessel function of the first kind of order 0 and argument β. The adapted phase deviation (β) may be determining by averaging the real and imaginary deviation of an OFDM signal, which can be represented by and a ₂ , respectively. Thereafter these real and imaginary OFDM deviations are scaled and referred to as the adapted phase deviation of the real and imaginary signal OFDM signal (β), this process can be represented as follows, where E[.], is the expected value ~ E[max(\ (m(t))\)] a ₂ ~ E[max(\3(m(t))\)] a

P, = T

a2

p, = -

According to another aspect of the invention, there is provided a method of controlling the PAPR of an OFDM signal in accordance with the method as described above, which further includes

an OFDM modulator, which subtracts the dominant frequency component from the modulated OFDM signal; and

an OFDM demodulator, which reinstates the dominant frequency component, in order to demodulate the phase modulated signal.

The method may include, an OFDM modulator adjusting the dominant frequency component control factor

The method may further include, an OFDM demodulator detecting the dominant frequency component control factor by examining the PAPR of the incoming signal, from which ίΙπβΨο,, ς and γ terms can be extracted by using a look-up table.

According to another aspect of the invention, there is provided a method of demodulating an OFDM signal, which includes

detecting a PAPR of an OFDM modulated signal; determining a dominant frequency component control factor ( γ), offset term (ψ ₀ ,) and a constant division term ( ς) of the OFDM modulated signal by examining the peak-to-average power ratio of the OFDM modulated signal; reintroducing the dominant frequency into the OFDM modulated signal; reconstructing the real and imaginary components of the OFDM signal; and

adding the real and imaginary components of the OFDM signal.

The method may include the further step of demodulating the OFDM modulated signal, by removing the high frequency components in order to obtain the difference between the scaled real OFDM, offset term and the scaled imaginary OFDM components, which can be given by The method may include the further step of demodulating the incoming

OFDM modulated signal, by multiplying it by a scaled phase modulated sinusoidal. Where the phase of the sinusoid is the scaled difference between the scaled imaginary OFDM components, the scaled real OFDM and offset term. This multiplication factor can be given by

The method may include the further step of demodulating the incoming OFDM signal, by removing the high frequency components in order to obtain the summation of the scaled real OFDM, offset term and the scaled imaginary OFDM components, which can be given by

ΦΛΨ

1 os +Φ 2.

The method may include the further step of demodulating the incoming OFDM modulated signal in order to obtain the real and imaginary OFDM components. The method may include the further step of obtaining a fast Fourier transform (FFT) of the OFDM signal.

The method may include the further step of passing the signal through an equalizer.

According to another aspect of the invention, there is provided an OFDM modulator, which, when operated, executes an OFDM modulation method as described above.

An OFDM modulator, which includes a scaling unit for scaling the imaginary and real components of the transformed input signal;

an offset term addition unit, for adding an offset term to the scaled real portion of the input message;

a phase modulator for modulating the scaled real and imaginary components of the input message;

a subtraction unit for subtracting the phase modulated OFDM imaginary component from the phase modulated OFDM real component; and

a subtraction unit operable to subtract a predefined dominant frequency from the signal.

According to another aspect of the invention, there is provided an OFDM demodulator which, when operated, executes a method of demodulating a modulated OFDM signal as described above.

According to another aspect of the invention, there is provided an OFDM demodulator, which includes

a PAPR detection unit for determining the PAPR of an input signal;

a lookup table referencing the PAPR against predefined values for a dominant frequency component control factor ( γ), an offset term (Ψ _^ ), and a constant division term ( g); a processor for determining the dominant frequency component from the values in the lookup table;

an addition unit operable to add a predefined dominant frequency to the signal; and

an OFDM demodulator for demodulating a modulated signal.

According to another aspect of the invention, there is provided a communication arrangement, which includes any one or both of a modulator as described above and a demodulator as described above. The invention will now be described, by way of example only with reference to the following drawing(s):

DRAWING(S)

In the drawing(s):

Figure 1 shows a schematic block diagram of a communication arrangement in accordance with one aspect of the invention;

Figure 2 shows an OM-OFDM modulator in accordance with another aspect of the invention;

Figure 3 shows an OM-OFDM demodulator in accordance with another aspect of the invention; and

Figure 4 shows the bandwidth occupancy of the communication arrangement of Figure 1 , when in use.

EMBODIMENT OF THE INVENTION

In Figure 1 reference numeral 10 refers to a communication arrangement, which includes an offset modulation orthogonal frequency division multiplexing (OM-OFDM) modulator 20, an offset modulation orthogonal frequency division multiplexing (OM-OFDM) demodulator 60 and a communication channel 100. The modulator receives an input message signal 1 1 at a transmitter end, this signal is modulated and after passing this modulated signal via the communication channel 100, this signal is feed into the OM demodulator which provides a reconstructed output message signal 13 at a receiver end.

The OM-OFDM modulator 20 is shown in more detail in Figure 2. At 12 the input signal is an N-Point inverse fast Fourier transform (IFFT). The IFFT is given by:

_N _ l .2Hk N i which can be considered to be an OFDM signal, where T is the symbol duration and represents the complex signal, which may also be written as aj+jb^. This signal may be modulated using the method described below.

At 22, by using a serial to parallel (S/P) converter, the incoming signal is separated into it's real and imaginary components.

The output of 22 produces real and imaginary components of the signal, which are then both scaled at 24 and 26 respectively, by dividing them with a constant scaling factor ς represented by:

where 21 and 3 refer to the real and imaginary parts of the OFDM message signal, ς refers to a constant division term, whereas Φ^(ή and Φ ₂ ( represent the equivalent scaled real and imaginary OFDM mapping. These O _j (t) and Φ ₂ ( components are interchangeable. At 28 a constant term Ψ is added to the real component of the signal.

At 30 (30.1 , 30.2, 30.3, 30.4), the adapted real and imaginary components are phase modulated. The difference between the respective phase modulated signals is taken at 32 (32.1 , 32.2).

At 34 (34.1 , 34.2) the baseband phase modulated signal is moved to a carrier frequency.

At 36 the resultant consinusoidal with the constant term is given by:

∞8(2π/ _£ ,ί+Φ ₁ ( +Ψ _Ο5 )- cos(2n _c t+O ₂ ( )- or represented as:

2s n sin ^ c ^t+ - where Ψ ₀₅ refers to the offset term. The parameters (Ψ , ς) are chosen such that^ »0 ₂ (t)-O _j (t), when Ψ ₀₅ is sufficiently large and Φ ₂ ( , ^φ ι( ^are sufficiently small. In this instance the Ψ term will dominate the expression.

At 38 the dominant frequency control factor given by,

( OS

2J ₀ (P) ² sinl 2 f _c t- I is subtracted from the combined signal before the signal is transmitted at 40 across the communication channel 100.

In Figure 3, the demodulator 60 receives the signal from the communication channel 100. The dominant frequency is reinstated at 63.

At 64 the received signal is squared producing a series of baseband and high frequency components which can be expressed as

1-cos

^cos ^20 ₂ +4n/Y)+ cos ^4π^ί+2Φ ₁ +2Ψ _ο ^.

At 66 the high frequency components of the squared signal are removed by using a low pass filter (LPF), the resultant signal can be given by

At 68 a constant term is added to the baseband signal, which can be given by

At 70 the inverse cosinousoidal process is performed producing

-Φ 2 +0> +Ψ .

1 os

At 72 an additional constant multiplication term is introduced resulting in

At 74 the signal is phase modulated by a sinusoid resulting in At 76 the signal is multiplied by a constant multiplication factor resulting in ^ Φ ₂ ( -Φ _{Ι (} -Ψ _Ο

^2sm { 2 j

At 78 the signal is transformed by a division process this results in At 79 the received input signal is multiplied by a sinusoidal carrier in addition to the term at 78, this results in

Φ _{Ι ( +} Ψ _Ο Φ ₂ (

sin[ 2 fj+ sin2n ^" t

At 80 the high frequency components are removed by using a low pass filter this results in a baseband signal represented by

At 82 a multiplication factor is introduced, this results in

At 84 an inverse co-sinusoidal operation is performed this results in

OS

At 88 a multiplication factor is introduced this results in

1 os 2

The subsequent steps at 90 (90.1 , 90.2), 92 (92.1 , 92.2), 94 and 96 (96.1 , 96.2) are used to isolate the real and imaginary OFDM components. The parallel to serial convertor (P/S) at 99 connbine these real and imaginary OFDM components.

The most prominent advantage of OM-OFDM is that by controlling the dominant frequency, the PAPR of the signal can be controlled. Without prior knowledge of the dominant frequency at the OM-OFDM demodulator, the dominant frequency can be determined by means of a look-up table which maps the PAPR of the signal against the other parameters. Therefore the dominant frequency can be determined without additional signal information. As an example the lookup table for a 16 quadrature amplitude modulation scheme is as follows:

At 63 to 99 the signal is demodulated with a uniquely developed OM-OFDM demodulation scheme and at 96 (96.1 , 96.2) the OFDM real and imaginary components are combined and presented at 99. In Figure 4, the bandwidth occupancy of the OM-OFDM signal is shown, as a spectrally efficient signal 1 12 with a dominant frequency component at 1 14. The dominant frequency component 1 14 is subtracted in the OM-OFDM modulator and reinstated in the OM-OFDM demodulator.

The inventors found that the OM-OFDM modulation method provides a spectrally efficient signal in which the PAPR can be adjusted without removing information from the signal and which does not lead to a severe bit error degradation. Furthermore, as the PAPR is used to determine the dominant frequency, no side information needs to be transmitted.

The inventors are of the opinion that the invention provides substantial advantages in the application of Digital Video Broadcasting (DVB), Worldwide Interoperability for Microwave Access (WiMax) and Long Term Evolution (LTE).