OF A WiBS ESS H ET OE

RELATED APPLICATION S Technical fiel of present disclosure may be related to United States Patent Application Mo, 12/752,471 entitled "Channel Quality

Estimation for MISE Receiver" filed on April 1 , ^' 201.0. The subjec matter of the related application is incorporated by reference in its entirety herein.

1 0

TECHNICAL FIELD

Technical field of present disclosure relates to adaptiveiy switching equalization operations m a node of a wireless network, and in particular to adaptivel switching between turbo equalisation operation I $ and other equalization operations based on expected performance of the turbo equalization operation.

BACKGROUND AND SUMMARY

In L -Term Evolution (LTE), single-carrier frequency-division 0 multiple-access (SC-F MAJ is se n the uplink, SC-PDMA is

advantageous in terms of power amplifier efficiency as it. has a smaller peak-to-average power ratio fPAPR) than an orthogonal f equency division multiple access (OPDMA) signal, SC-FDMA, however, gives rise to an inter-symbol interference (IS!) problem in dis e siv channels, it is important to address IS! so that SC--FDMA ca m r ve power ¾m «.r eflciency witho t sacx ilcm p formimce>

When LTE is first rolled out, it is likel that linear minimum mean square error fL MSB) receivers will be implemented in the base S station, also referred to as an e odeB, LMMSE receivers suppress ISI using linear frequency-domain equalisat on, where the filter coefficients are designed o ma i ze the si^3al~ o-in ier ere ce- .hjs-nmse ratio {SINK} for each subcarrier com onent. Compared to a almple matc filtering rece ve , LMMSE provides a significant performance

0 improvement.

But it is thought that even better performance can be achieved by employing even more advanced -receiver techniques. For example, there have been interests in using a turbo- receiver for turbo equalizer) in uplink LTE to improve performance in i$!. channels beyond LMMSE. S However, the complexity of a turbo receiver is high. Thus, it is

advantageous to turn on the iterative turbo processing only when there is a good chance of performance improvement from the turbo

processing.

A method has been, proposed to adoptively switch on and off the0 iterative turbo operation. This previous method is based on the

post-equalkation Si. E¾ of an .MSB receiver and a turbo receiver. Comparing these two SINR gi es ise to a gain factor ( In the calculation of the turbo receiver SINK, it is assumed that the ISi in the turbo receiver is completely removed. The previou .method further depends on an estimated average bit error rate BER) indic tor, .

Whether the iterative turbo operation is activated or not is determined based on G and Έ ,

While the prev o s method has boon shown to be effective, a even better solution may be achievable through estimating performance ^• using, for example, a capacity-achieving receiver,

SUMMARY

A non-limiting aspect of the discl sed subject matter is directed to method to adoptively switch equalisation operati ns in a wireless _* network- The method may be performed at. or on behalf of a node of the wireless network. The method comprises receiving irelessly a signal from a transmitter; estimating a post-equalisation S1 pe formance of a turbo equali ation operation based, on the received, signal; and selecting, based on the estimated post-equallaation S!N _. performance of the turbo equalisation operation, one of a turbo eqnali&ation opera tion and a linear e u lisati n operation to equali e the received atgnaL The proces to estimate the po¾t-cqnali»ation SINR performance the turbo equalisation operation comprises determining per-snbeamer SI ^' Rs for a plurality of suhcarrier allocated to the received signal; determining per- subcarrier capacities for the suhcarriers allocated to the received signal based on the per~snbearrier Sl ^' MRs; and estimating a. post-equalimtion SINR performance of the turbo equalization operation based on th per-suhearrier capacities of the subcarriers. Another nan-limiting aspect of the disclosed subject matter is directed to a node of a reless network structured to dopt vely switch equalisation operations. The node compris s a turbo receiver structured to perform a. turbo equalisation operation on a received S signal; a linear receiver structured to perform a linear equalisation

operation on the received ignal; a turbo performance estimator structured to estimate a posi~e uali;satksn S performance of the turbo receiver based on the received signal; auo! a switch controller structured to switch the received signal to be equali e , by one of the 0 turbo receiver and the linear receiver based on the estimated

post-cquall¾atio«. SINE p f rmance of the turbo receiver. The turbo performance estimator i structured to estimate the post-equalization SI MR performance of the turbo receiver by determining per-suhc&rrier SlNRs for a plurality of subcarriers allocated to the received signal, S determining per-s bearrier capacities for the subeamcrs allocated to the received signal based on the pet- anticarrier SiHRs, and estimating a post-equalisation SINE performance of the turbo receiver based on the per- siifaearrier capacities of the aubcarders.

Yet another non-limiting aspect of the disclosed subject matter is0 directed to a uou-transitory computer storage medium which has

stored therein programming instructions. When a computer e ecu es the programming instructions, the computer executes a method to adaptlvely switch qualisation operations in a wireless network, The me thod may foe performed at or Oft behalf of a node of the wireless network. The method comprises receiving irelessh _/ ' a signal, from a transmitter; estimating a osi-e u&Ik^u xi SI E performance of a turbo equaMm fou operation based, on the received s n l; and selecting, based on the estimated, poshequalixatio SI R performance of the turbo e ualisation operation, one of a turbo equ l sati n operatio and a linear equalimtion operation to equalise the received signal. he process to estimate the os -e alisation SI E performance the turbo equalisa ion operation comprises cleterminrng per-su camer SiJNEs for a plurality of subcarriers allocated to the received signal; determining per-suhearrie capacities for the sufocarriers allocated to the received signal based on the per-snbeamer SINEs; and estimating a p st-equaliaation SINR performance of the turbo equalisatio operation based on the per-subcarrier capacities of the snhcaraers. DESCRIPTION OF THE DRAWINGS

The foregoing and other objects,, features, and advantages of the disclosed subject matter will be apparent horn the .following more particular description of preferred embodiments as illustrated In the accompanying drawings in which reference characters refer to the same parts throughout the various vi ws. The drawings are not necessarily to scale.

Fig, 1 illustrates an example scen rio of a wireless network in which mobile terminals and base station communicate with each other; Fig. 2 illustrates an example embodiment of a node of a wireless network; nd

Fig. 3 illustrates an ex mple method to ac!aptivei switch equalization operations;

Fig, illustrates an ex mpl process to estimate a

post-equa.H:«aiion SI R performance of a turbo receiver;

. S illustrates an exam le process to estimate a

post-eqnaliaauou SINR performance baaed en per-subcarrfer capacities;

Fig. <> illustrates an example process to estimate a

posi ~equ.ali2B.tion SINR performance of a linear receiver;

Figs. ?, 8 and 9 illustrate example rocesses to switch received signal to be equalised between turbo and linear receivers; and

Fig, 10 illustrates an example embodiment of a turbo receiver in a node of a wireless network,

DETAILED DESCRIPTION

For purposes of explanation and not limitation., specific details are set forth such as particular architecture , interfaces, technlo-iies, and so on. However, it will be apparent to those skilled in the art that d e technology described herein may be practiced in other

embodiments that depar from these specific details. That, is, those skilled in the art will be able to devise various arrangements which, although not explicitly described or shown herein, embody the principles of the described technology. In some instances,, detailed descriptions of well-known devices, circuits, and methods are omitted s s not to scur the description with unnecessary details. AH s a em ts herein reciting principles, aspects, embodiments and examples are intended to encompass both S structural and functional equivalents. Additionally, it is intended that such equivalents include both currently known equivalents as well as equivalents developed in the future, i.e., any elements developed that perform same function, regardless! of structure.

Thus, for exam le, it will he appreciated that block diagrams0 herein can represent conceptual views of illustrative circuitry

embodying principles of the technology. Similarly, it will he appreciated that any flow charts, state transition diagrams, pseudo code, and the like represent various processes which may be substantially

represented in computer readable medium and executed by a computer S or processor, whether or not such computer or processor is explicitly shown.

Functions of various elements including functional blocks labeled or described as "processors" or "controllers" may be provided through dedicated hardware as well as hardware capable of executing

0 associated software. When provided by a processor, functions may be provided by a single dedicated processor, by a single shared processor, or by a plurality of individual processors, some of which may be shared or distributed. Moreover, explicit use of term " rocessor" or "controller ^* should not be construed to refer exclusively to hardware capable of executing software, and may include, without limitation, digital signal rocessor {s or ened to ^* D$P ^* } ardware, ro¾d only memory {shortened to "ROM") for storing software, random access memory {shortened to RAM , and nonvolatile storage.

in this doc men , 3GPP, and in particular VTE _f is primarily used as examples for explanation purposes. Note th the technology described herein can he applied to :aon~3GPP standards such as cdma20O0 ₅ EV-DO, TD-SCD A, or other 3 GPP stand rds such as WCDMA and HSFA, Thus, the scope of this disclosure is not limited to the set of 30PP wireless network systems and can encompass many domains of wireless network systems. Also, a base station fe,g> H'BS, NodeB, eNodeB, eNB, etc.) will be used as an example of a network node in which the described method can be performed. However, It should foe noted that the disclosed subject matter is applicable to any node, such as relay stations, that receive wireless signals. Also without less of generality, mobile terminals fe,g, HE, mobile computer, PDA, etc, will be used as examples of wireless terminals that communicate with the base station.

Fig. I. illustrates a ex m le scenario of a wireless network 100 in which mobile terminals 130 and a base station. 1 10 communicate with each other. At the base station 110, signals transmitted from any mobile station 130 are equalised to increase the eileet ?e S1NR of th communication link between die base station 110 and the mobile terminal 130. In this scenario, the base station i 10 as the receiver of the s nals transmitted iron* the transmitters (the mobile terminals 130) pe f ms t e equalisation of the signals. However, this la not a.

limitation. AIT network node may be able to perform such

equaliza ion.. Indeed, in the downlink direction, it. is contemplated that some or all mobile terminals 130 may also perform e ualisation processes to which one or more aspects of the disclosed subject, matter are applicable. For ease of e lanati however, ft is assumed t ai the equalisation is performed at the base station 1 10- Fig. 2 illustrates an example embodiment of a base station 10 of a wireless network KM).. As seen, the example base station 1 10 includes a turbo receiver 210, a turbo performance estimator 220, a linear receiver 230, a linear erformance estimator 240, a switch controller 250, and a switch 2 ί>, The linear performance estimator 240 is shown as a dashed bo . to indicate that it is optional.

These units of the base station 1 10 are structured to adaptlveiy switch equalisation operations performed on the incoming signal. Since the base station 1 10 is but one of several nodes of the network 100 that can perform the adaptive switching operations;, the terms "n ^* , and "base station ^* and variance thereof (e.g.. Mod B, e!fede D, el^B, etc.), may be used interchangeably. Thus, Fig.. 2 may be viewed a an embodiment of a node 1 10 capable of performing adaptive switching operations. It should be rioted that Fig, 2 is a logical representation of he node 110, Thus, each of the units (the turbo receiver 210, the turbo performance estimator 220, the linear receiver 230, the linear performance estimator 240 _» the switch controller 250, the switch 255), need not be physically separate from an of the other units. It is folly contemplated that any combination of the units may be integrated into a single physical device. Further, any of the u s may be hnplemeni d. in. multiple physical components operativeiy structured and coupled to each other to perform the respective function of the -unit Yet further, to the exten that, some of the units share common features, multiple units ma share common components.

While not explicitly shown. It is also contemplated, that the node 110 as a whole can be implemented as a combination, of hardware and software components. For x m le, the node 1 10 may include one or more processors, which as described above can he combinations of hardware and software arranged to perform the functions associated with the units, The processor^! may execute programming

instructions stored in a no -transitory computer-readable medium to perform the functions. The programming instructions m y also be received in a transitory manner and stored in the non- transitory computer-readable medium accessible to the node 1.1.0.

Fig, 3 illustrates an example method to adsphvely switch equalisation operations. This method may be performed at or on behalf of the node ! 10 of the wireless network- From one perspective, the I I method 300 may be described as capturing the expected gain of the turbo receiver 2 0 through an SIMR e ess of a capacity achieving receiver. This way of capturing the ost- ualiz SINE may better characterise the actual performance of the turbo ec|ua!i¾atiou operation performed by the turbo receiver 210, As a consequence,, better decisions may be made in determining whether the iterative turbo opera ion should be act va e or not.

The method 300 begins in step 310 when a signal, from a.

transmitter 130 is wixelessfy rec ed. Based on the received signal, the turbo performance estimator 220 estimates a post-equalization SIMR performance _y of the turbo equalisation operation performed by, for example, the turbo receiver 2 Ϊ0 in step 320, In step 340 _» the switch controller 2S0 switches or selects the received signal to he equalized b eithe the turbo receiver 2.10 or the linear receiver 230 based the estimated post- equaikabon Si$R performance y _r of the turbo receiver 210. Optionally, the linear performance estimator 240 may estimate the post-equalization SIMR performance of the linear equalization operation performed by, for example, the linear receiver 230, and the switch controller 250 may select or switch the received signal also based on the estimated performance of the linear receiver 230. Thus, in one embodiment, the switching decision made by die switch controller 250 is based on the estimated performance y _r of the turbo receiver 10 alone, and m another embodiments, the switching decision is based cm the estimated performances γ _τ , f _M of both the turbo arid linear receivers 210, 230-

The terms '"switch" and "switching" should not be taken to be limiting. That is, the terras are broader than sim l operating the switch 255 to multiplex the outputs of the turbo receiver 210 and the linear receiver 230 based on the es imated performances as illustrated in Fig. 2< Rather, th se te ms re intended to be interpreted broadly to incorporate the concepts of "activating", "en bling , "causing",

"selecting/' and so on. In other words, in step 340, the switch

10 controller 250 takes actions to ensure that he proper equalisation operahon is performed on the received signal For ex m le, in

particular embodiments, switch controller 250 nd switch 255 ma represent a processor or multiple processors capable of selecting an appropriate equalisation operation o perform or of selecting for further

I S ^• use one the ou uts generated by multiple differen eq alk&tion

operations that are performed in parallel on received signals.

Fig. 4 illustrates an example process to implement step 320 performed y the turbo performance estimator 220 to estimate the post-eqnalisaho t SINE performance _r of the turbo receiver 210, In step 10 of the process, the per-subearrler SINRs m determined tor a plurality of subcarriers allocated to the received signal lu a non- limiting aspect, a per-suhcarrier SINE is determined, as follows. .

In equation. (1), iNR^. deno es he per-so&carrier SI.NR es im t , B, denotes the symbol energy, jV _¾ denotes the one-sided hit noise power spectral density, H[k] denotes the d s ersive channel response for the su eam r, and k denotes the snbcarriet Index.

in step 420, the turbo performance estimator 220 determines the per-sinbearrier capacities for each of the subearriers. The

per--subcarrkr capac ty may be termined as follows _*

< \ ^ logs ! ÷ .¾ ¾ ) |2)

The log in equation (2) ma comprise base 2 log calculations, or may comprise other base log calculations or natural log,

h step 430 _» the turbo performance estimator 220 estimates the post-ecjuall¾ tion SINE performance y _r . of the turbo receiver 210» Fig. S illustrates an example process to im lement the step 430. First, m step 510, the turbo performanee estimator 220 sums and averages the per~suhearrier capacities are to compute an average capacity C _: . as follows.

In equation |3K K is the number of subearriers in the received signal.

The average capacity C _mi is used in step 520 to estimate the

SIMR -performance _r of the turbo receiver 210 as follows. in equation 4} _r natural logarithm is assumed to be used in. the capacity ¾ computation. hus, the exponential num er m used as the base when converting the average capacity€ _Am to the post-eqttaikation SINR in e ion (4) . If other bases are used in the capacity

computation. C _t of the individual suhearriers, qua on (4| should be dj sted accordingly.

By combining eq a ions {I ^' H^h th« post-equalkatiou SINE f _r is given by equation fS],

equation (5!,. M represents a number of recei e antennas, and #J¾J represents the frequency response of the channel corresponding to the A* subcarrier from the transmit antenna to the receive antenna.

Generally, noise and interference can exhibit correlation across frequencies and receive antennas. If RJk ^' l denotes an MxM correlation matrix corresponding to the * ^w subcarrier and Hi ¾ j repr s nts vector collecting the frequency responses of the subcarrier from the transmitter to all M receive antennas, then the per~subearriet SINR determined s follows.

in this equation, the RJk) matrix also referred to s the impairment correlation .matrix corresponding to the subcarrier, In a ition, H^ ^' f&j re resents the Rermitian transpose of the vector 8{ ^' £ j .. The por-su ^' bcamer capacity and average capacity n still e

determined usi g equations (2) and (3), respectively- Then equation (4) can be used to determine the post«equali¾atfon BJNR o the urbo receiver 210.

Referring back to Fig. 3, it has been m ntioned that in one non-limiting aspect, a comparison is made between the estimated performances of the turbo and linear eceiv rs 210 and 230 y _r a d _5i and switching decision in step 340 is made based on the comparison, If a linear minimum mean square erro fLM SE) receiver 210 is assumed, the the post-equalization S1HR performance for the LMMSE .receiver 210 can be estimated by as follows.

F¾ . 6 illustrates an example process performed by the linear performance estimator 240 to perform step 330 for estimating the post-equalisation SINE performance y of the linear receiver 230. In step 610, the linear performance estimator 240 determines

per-subcarriers SLMHs similar to step 410 performed by the turbo performance estimator 220. in step 620, the linear performance estimator 240 determines a mean reliability of the subcarriers.

In. equation (7), the estimated performance y _M of the linear receive 230 is based on a harmonic mean of the uan ity (1 SfNH^) . From one perspective, this -q ant ty can. be described as quantitative m sure of how m c m&rmatiosi c n fee reliably transmit ed in otie symbol over the s bcamer For example, over a very low noise sttbc4¾me-r _> Le. _.> » 1 , a very high modulation such as 64-QA may be used. This translates to 6 bits per symbol. Over a moderately tew noise s focarrier, 16-QAM may foe used, meaning that 4 bits can be transmitted in one symbol duration. Note that even m a very noisy environment sometimes described as a pou r- mited. regime, i.e., $iA ^' *0 _> some information can still be transmitted {e.g., 1 bit per symbol). Of course,, the actual amo of data that can foe transmitted will also depend on the coding scheme.

When the quantity (1 * Sih ) is high, more informati n can be sen reliably in one s mbol over the aubearrier. Conversely , when the quantity (1 Si 'R* ) is low, the amoun t of Information that can he sent In one symbol ove the sufocarrier is reduced, That , each (I SINR _k ) quantity represents a quantitative measurement of the sub-carriers reliability, in. equation (?} ₅ a mean (harmonic mean in. this instance) of these reliability measummenta Is expressed. Thus, in step 630» it can be said that the linear performance estimator 240 determines the estimated performance y _¾i based on a mean of quantitative reliability measurements of the sube&rriers,

Mote that the mean of reliability measurements is not limited to the harmonic mean. Other ways to determine central tendencies such I? as arithmetic mean or geometric mean may be «se L In this instance,, &rmomc mean, k sed since it is better suited, for determini g central, tendencies when ratios are involved.

Going back to fig, 3, the switch, controller 250 switches the received signal to be equalised by one of the turbo receiver 210 and the linear receiver 230 based on. the Slr!E performances ? _T and y _M i step 340,

Fig. 7 illustrates an example process performed, b the switch, controller 250 to implement step 340 when the estimated SI E performances y _r of the turbo receiver 210 and. the linear receiver 230 ¾ _i are available. In step 710, the switch controller 250 calculates a gain atio _r as follows.

It should be noted that the gain rado G _r of equatio (9) merely one of several ways that the estimated perfor aaces y _r and y - _M can foe related. For example, the gain m y be expressed as a difference rather than as a ratio. Then n step 420, the switch controller 250 determines whethe the an ratio <¾. is greater than a predetermined minimum ratio th shol - When <* _? is greater than , the received signal is switched to he equalised by the turbo receiver 210 i step 430- if on the other hand the gain ratio ϋ _τ i& not greater than _w , the switch controller 250 switches the received signal to be equalised by the linear receiver 230 in step 440. As men ioned above,, the term "sw tch ^* should be taken m the broad sense.

But in another non-limiting aspect, the decision as to whether the .received signal should he equalized hy the turbo receiver 210 or the linear receiver 230 may be made based solely on the post-equa!i¾aho« SI R performance γ _γ of he turbo receiver 210 est ma ed, by he t rbo performance estimator 220 in step 320. In other words, the estimated performance of the linear eceiver 230 need, not be considered.

Fig. 8 illustrates a rocess to perform step 340 without consideration, of the estimated performance ?. of the linear receiver

230, In step 810 of the process, the switch controller 250 est mates a required S R performance modulating encoding scheme

I ^' CS) of the received signal, la step 820, the switch controller 250 determines whether or not the estimated performance of the turbo receiver 10 is greater than required SINR ? _{m m} of the received signals MSC In step S30, if the estimated performance ,f _r is

determined to be greater than the required SINR y _{m im#} of the received signal's .CS, the switch controller 2 SO switches the received signal to he equaii¾ed by the turbo receiver 210 m step 830, Otherwise, the received signal is switched to be e n iked by the linear receiver 230 in step 340.

Fig. 9 illustrates another exam le process to switch the received signal to be equalised between the turbo and linear receivers 210 _? 230. The process illustrated in Fig. 9 Is explained as follows. If the expected performance of the turbo receiver 210 is lower tha,n the re uired S!NR lor the receive signal's MCS, there is no expected benefit from ttsing the turbo operation over the linear operation for e u l sation, even if the expected gain of the turbo ecei er 2.1.0 over the linear receiver 230 is large. In other words, if y _r Is less than s x mw _^ whether 6 i large or small can become irrelevant.

Thus, in Fig. 9, the sw tch controller 250 in steps 910 and 920 determines the required 8IMR r - _® ^ ^CS of the rec ived signal, and. dete m es whether or not the estimated post-equaM¾a ion SI R perforro.an.ee y _r of the turbo receiver 210 is greater than y _{9iiiri l9Vfs} . These steps are similar to steps 810 and 620 illustrated in Fig. 8. If it Is determined In step 920 that die estimated performance y _T is good enough, le,. It is greater than ?» υ _® *α _* the estimated performance of the linear receiver 230 is then considered before making the final switching decision. If the estimated SINR performance y Is net good enough, Le., y _T Is not greater than ^ ^ ₉ the in step 900, the switch controller 250 switches the received signal to he equalised by the linear receiver 230,

If in step 920 the estimated SINR performance y _? of the turbo receiver 210 Is greater than the y _{m ;} - _fi , then the switch controller 250 calculates the gain ratio 0- _f in step 930, In step 940, the switch. controller 250 determine whether or not die gain ratio C¾- is greater than the ed termined minimum gain ratio threshold C? _¾¾¾ , , If the gain ratio Cir is greater than the threshold 0 _ΑίίΧ , then in ste 950, the switch controller 2S0 switc es the received signal to be e ualled y the turbo receiver 210- Else, the rec i ed signal is switched to be qualized by the linear receiver 230 n step 960,

Of th three implementations, illustrated in Figs. ?, 8 and 9, the process in. Fig, 9 is the most intelligent

Fig, 10 illustrates a exampl embodiment of the turbo receiver 2.1.0. As seen, the turbo receiver 210 comprises a. minimum, mea square error decision feedback equalker ( MSB-DFE) 1010 which outputs an equalised signal based o.n the received signal, a

desera-mbier deinterleaver 1.020 structured to outpu a deserambkrd and demter!eaved signal from, ie equalised signal, a decoder 10 0 structured to output a decoded signal from the deserarnblecl a»d deinteriea.ved sigs al, a scramb!er/lntcrleave .1.050 structured to output a. scrambled and interleaved signal, from the decoded signal, and a soft symbol modulator 1060 structured to output soft symbols, which are estimates of the transmi ted symbols, from the scrambled/interleaved signal The soft symbols are provided to the MMSE-DFE 1010.

The advantage of this turbo receiver 210 is that, it can perform both turbo equalisation operations as well, as linear equali ation operations. The turbo receiver 21.0 performs turbo equalis ion operations when e switch 1055 is closed to form a feedback loop, if the switch iOSS is o ened to thereby discon ec the feedback k>op _> the MMSE-DFE 1010, the descram ler/diiinterkayer 1020 and the decoder 1040 cooperatively function together as the linear receiver, In other words, th output of the decoder 1040 with feedback to the

scrambler / mterle&ver .1050 is e uivalen to the received signal being equalised by turbo e ualization, and the outp of the decoder 1040 without feedback to the & mmhler/mt©rk¾ver 1050 is equivalent to the received signal being equalised by linear equalisation,.

in this embodiment of th node HO, the switch controller 250 opens and closes the switch 1055 of the turbo .receiver 21(1 to switch, the received signal for either the turbo or linear equalisation operation. Again, it is emphasised that switching is to he broadly Interpreted, Mole that switching in this instance can he thought of as activating a. turbo mode or linear mode.

Due significant advantage of the disclosed subject matter is that the complicated iterative turbo operation can he accurately turned off when the expected _. erformance benefit is small or none.

Although the description above contains many specificities _* these should not he construed as limiting the scope of the disclosed subject m ter but as .mere iy providing illustrations of some of the presently preferred mbodiments. Therefore, It will be appreciated, that the scope of the disclosed subject matter fully encompasses other embodiments which may become obvious to those skilled in die art, and that the scope is accordingly not to bo limited, AH structural, and functional equivalents to the elements of the shove^lescril d. preferred

em d ment that are kno n to those of ordinary skill in the art are ex ressl incorporated herein by reference and are intended, to he encompassed hereby. Moreover, it is not necessary for a device or method to address each and every problem described herein or sought to be solved by the present technology, for it to be encompassed ^' hereby.