VENSOFT Technologies www.ieeedeveloperslabs.in Email: vensoftprojects@ ahoo.com !

ontact: "##$$#%$%#

&EEE '()* +,T-,. /0O1E!TS ,!,2E+&! 3E,0 '()*4'()# FO0 +.Tech5 ..E5..Tech '(. /hase Noise in +&+O S stems: .a esian !ram6er40ao .o7nds and Soft4&np7t Estimation Abstract: This paper addresses the problem of estimating time varying phase noise caused by imperfect oscillators in multiple-input multiple-output (MIMO) systems. The estimation problem is parameterized in detail and based on an equivalent signal model its dimensionality is reduced to minimize the overhead associated ith phase noise estimation. !e e"act and closed-form e"pressions for the #ayesian $rame%r-&ao lo er bounds (#$&'#s) and soft-input ma"imum a posteriori (M()) estimators for online* i.e.* filtering* and offline* i.e.* smoothing* estimation of phase noise over the length of a frame are derived. +imulations demonstrate that the proposed M() estimators, mean-square error (M+-) performances are very close to the derived #$&'#s at moderate-to-high signal-to-noise ratios. To reduce the overhead and comple"ity associated ith trac.ing the phase noise processes over the length of a frame* a novel soft-input e"tended /alman filter (-/0) and e"tended /alman smoother (-/+) that use soft statistics of the transmitted symbols given the current observations are proposed. !umerical results indicate that by employing the proposed phase trac.ing approach* the bit-error rate performance of a MIMO system affected by phase noise can be significantly improved. In addition* simulation results indicate that the proposed phase noise estimation scheme allo s for application of higher order modulations and larger numbers of antennas in MIMO systems that employ imperfect oscillators. /7blished in: +ignal )rocessing* I--- Transactions on (1olume234 * Issue2 45 ) &ss7e 2ate : May46* 7548 Inde" Terms9Multi-input multi-output (MIMO)* :iener phase noise* #ayesian $ram;er &ao lo er bound (#$&'#)* ma"imum-a-posteriori (M())* softdecision e"tended /alman filter (-/0)* and e"tended /alman smoother (-/+). '). +7lti scale 8ossip for Efficient 2ecentrali9ed ,veraging in :ireless /ac;et Networ;s Abstract: This paper describes and analyzes a hierarchical algorithm called Multi scale <ossip for solving the distributed average consensus
VENSOFT Technologies www.ieeedeveloperslabs.in Email: vensoftprojects@ ahoo.com !ontact: "##$$#%$%#

VENSOFT Technologies www.ieeedeveloperslabs.in Email: vensoftprojects@ ahoo.com !ontact: "##$$#%$%#

problem in ireless sensor net or.s. The algorithm proceeds by recursively partitioning a given net or.. Initially* nodes at the finest scale gossip to compute local averages. Then* using multi-hop communication and geographic routing to communicate bet een nodes that are not directly connected* these local averages are progressively fused up the hierarchy until the global average is computed. :e sho that the proposed hierarchical scheme ith k=>(loglogn) levels of hierarchy is competitive ith state-of-the-art randomized gossip algorithms in terms of message comple"ity* achieving ?-accuracy ith high probability after O(n loglogn log@4A(?)B ) single-hop messages. /ey to our analysis is the ay in hich the net or. is recursively partitioned. :e find that the above scaling la is achieved hen sub net or.s at scale j contain O(n(7A8)j) nodesC then the message comple"ity at any individual scale is O(n log@4A?B). (nother important consequence of the hierarchical construction is that the longest distance over hich messages are e"changed is O(n4A8) hops (at the highest scale)* and most messages (at lo er scales) travel shorter distances. In net or.s that use lin.-level ac.no ledgements* this results in less congestion and resource usage by reducing message retransmissions. +imulations illustrate that the proposed scheme is more efficient than state-of-the-art randomized gossip algorithms based on averaging along paths. /7blished in: +ignal )rocessing* I--- Transactions on (1olume234 * Issue2 D) 2ate of /7blication: May4* 7548 ''. 1oint Estimation of !hannel and Oscillator /hase Noise in +&+O S stems ,bstract: Oscillator phase noise limits the performance of high speed communication systems since it results in time varying channels and rotation of the signal constellation from symbol to symbol. In this paper* Eointestimation of channel gains and :iener phase noise in multi-input multi-output (MIMO) systems is analyzed. The signal model for the estimation problem is outlined in detail and ne e"pressions for the $rame%r&ao lo er bounds ($&'#s) for the multi-parameter estimation problem are
derived. ( data-aided least-squares ('+) estimator for Eointly obtaining the channel gains and phase noise parameters is derived. !e"t* a decisiondirected eighted least-squares (:'+) estimator is proposed* here pilots and estimated data symbols are employed to trac. the timevarying phase noise parameters over a frame. In order to reduce the
VENSOFT Technologies www.ieeedeveloperslabs.in Email: vensoftprojects@ ahoo.com !ontact: "##$$#%$%#

VENSOFT Technologies www.ieeedeveloperslabs.in Email: vensoftprojects@ ahoo.com !ontact: "##$$#%$%#

overhead and delay associated ith the estimation process* a ne decisiondirected e"tended /alman filter (-/0) is proposed for trac.ing the MIMO phase noise throughout a frame. !umerical results sho that the proposed '+* :'+* and -/0 estimators, performances are close to the $&'#. 0inally* simulation results demonstrate that by employing the proposed channel and time-varying phase noise estimators the bit-error rate performance of a MIMO system can be significantly improved.

/7blished in: +ignal )rocessing* I--- Transactions on (1olume235 * Issue2 D) 2ate of /7blication: +ept. 7547 Index Terms$hannel estimation* $ramFr-&ao lo er bound ($&'#)* e"tended /alman filter (-/0)* multi-input multi-output (MIMO)* eighted least squares (:'+)* :iener phase noise. '*. ,cc7rate !omp7tation of the +8F of the -ognormal 2istrib7tion and its ,pplication to S7m of -ognormals ,bstract: +ums of lognormal random variables (&1s) are of ide interest in ireless communications and other areas of science and engineering. +ince the distribution of lognormal sums is not log-normal and does not have a closed-form analytical e"pression* many appro"imations and bounds have been developed. This paper develops t o computational methods for the moment generating function (M<0) or the characteristic function ($G0) of a single lognormal &1. The first method uses classical comple" integration techniques based on steepest-descent integration. The saddle point of the integrand is e"plicitly e"pressed by the 'ambert function. The steepest-descent (optimal) contour and t o closely-related closedform contours are derived. ( simple integration rule (e.g.* the midpoint rule) along any of these contours computes the M<0A$G0 ith high accuracy. The second approach uses a variation on the trapezoidal rule due to Ooura and Mori. Importantly* the cumulative distribution function of lognormalsums is derived as an alternating series and convergence acceleration via the -psilon algorithm is used to reduce* in some cases* the computational load by a factor of 453H Overall* accuracy levels of 48 to 46 significant digits are readily achievable. /7blished in: $ommunications* I--- Transactions on (1olume26I * Issue2 6) 2ate of /7blication: May 7545 Index Terms<+um characteristic function. of lognormals* moment-generating function*

VENSOFT Technologies www.ieeedeveloperslabs.in Email: vensoftprojects@ ahoo.com !ontact: "##$$#%$%#

VENSOFT Technologies www.ieeedeveloperslabs.in Email: vensoftprojects@ ahoo.com !ontact: "##$$#%$%#

'#. !ompressed Sensing of EE8 for :ireless Tele monitoring with -ow Energ !ons7mption and &ne=pensive >ardware ,bstract: Tele monitoring of electroencephalogram (--<) through ireless body-area net or.s is an evolving direction in personalized medicine. (mong various constraints in designing such a system* three important constraints are energy consumption* data compression* and device cost. $onventional data compression methodologies* although effective in data compression* consumes significant energy and cannot reduce device cost. $ompressed sensing ($+)* as an emerging data compression methodology* is promising in catering to these constraints. Go ever* --< is non sparse in the time domain and also non sparse in transformed domains (such as the avelet domain). Therefore* it is e"tremely difficult for current $+ algorithms to recover --< ith the quality that satisfies the requirements of clinical diagnosis and engineering applications. &ecently* bloc. sparse #ayesian learning (#+#') as proposed as a ne method to the $+ problem. This study introduces the technique to the tele monitoring of --<. -"perimental results sho that its recovery quality is better than state-of-the-art $+ algorithms* and sufficient for practical use. These results suggest that #+#' is very promising for tele monitoring of --< and other non sparse physiological signals. /7blished in: #iomedical on (1olume235 * Issue2 4 ) 2ate of /7blication: Jan. 7548 Index Terms9Telemonitoring* Gealthcare* :ireless #ody- (rea !et or. (:#(!)* $ompressed +ensing ($+)* #loc. +parse #ayesian 'earning (#+#')* electroencephalogram (--<) '?. !ompressed Sensing for Energ 4Efficient :ireless Tele monitoring of Non4&nvasive Fetal E!8 via .loc; Sparse .a esian -earning ,bstract: 0etal -$< (0-$<) tele monitoring is an important branch in telemedicine. The design of a tele monitoring system via a ireless body area net or. ith lo energy consumption for ambulatory use is highly desirable. (s an emerging technique* compressed sensing ($+) sho s great promise in compressingAreconstructing data ith lo energy consumption.
VENSOFT Technologies www.ieeedeveloperslabs.in Email: vensoftprojects@ ahoo.com !ontact: "##$$#%$%#

-ngineering*

I---

Transactions

VENSOFT Technologies www.ieeedeveloperslabs.in Email: vensoftprojects@ ahoo.com !ontact: "##$$#%$%#

Go ever* due to some specific characteristics of ra 0-$< recordings such as non sparsity and strong noise contamination* current $+ algorithms generally fail in this application. This paper proposes to use the bloc. sparse #ayesian learning frame or. to compressAreconstruct non sparse ra 0-$< recordings. -"perimental results sho that the frame or. can reconstruct the ra recordings ith high quality. -specially* the reconstruction does not destroy the interdependence relation among the multichannel recordings. This ensures that the independent component analysis decomposition of the reconstructed recordings has high fidelity. 0urthermore* the frame or. allo s the use of a sparse binary sensing matri" ith much fe er nonzero entries to compress recordings. )articularly* each column of the matri" can contain only t o nonzero entries. This sho s that the frame or.* compared to other algorithms such as current $+ algorithms and avelet algorithms* can greatly reduce code e"ecution in $)K in the data compression stage. /7blished in: #iomedical on (1olume235 * Issue2 7 ) 2ate of /7blication: 0eb. 7548 Index Terms<0etal -$< (0-$<)* Tele monitoring* Telemedicine* Gealthcare* #loc. +parse #ayesian 'earning (#+#')* $ompressed +ensing ($+)* Independent $omponent (nalysis (I$() -ngineering* I--Transactions

VENSOFT Technologies www.ieeedeveloperslabs.in Email: vensoftprojects@ ahoo.com !ontact: "##$$#%$%#