Skip to main content

Ramesh Annavajjala

Affiliate Research Associate Professor

Contact

Er.annavajjala@northeastern.edu

Research Interests

  • Wireless Networks
  • Machine Learning and Signal Processing
  • Software Defined Networking
  • Communication and Computing

Education

  • PhD in Electrical Engineering, University of California – San Diego
  • MS in Telecommunications, Indian Institute of Science – Bangalore, India
  • BS in Electronics and Communication Engineering, National Institute of Technology – Warangal, India

Biography

Ramesh Annavajjala is an Affiliate Research Associate Professor at Northeastern University’s College of Computer and Information Science. He received his Bachelor’s Degree in Electronics and Communication Engineering from the National Institute of Technology (NIT, Warangal, India, May 1998), his Master’s Degree in Telecommunications from the Indian Institute of Science (IISc, Bangalore, India, January 2001), and his PhD in Electrical Engineering from the University of California at San Diego (UCSD, La Jolla, CA, June 2006.)

Dr. Annavajjala has served as a Distinguished Member of Technical Staff at the Altiostar Networks Inc., a Tewksbury, Massachusetts-based startup company, developing cloud-RAN optimized 4G LTE base-station products. Prior to that, he was a Principal Member of Research Staff at the Mitsubishi Electric Research Labs (MERL), in Cambridge, MA. He has also held industry positions at ArrayComm LLC (San Jose, CA, USA), Synopsys Inc., (Bangalore, India) and CMC R&D Center (Hyderabad, India).

Dr. Annavajjala received the Fred W. Ellersick MILCOM award for the “Best Paper in the Unclassified Technical Program” for his paper titled “Design and Performance Evaluation of High-Capacity Mobile Troposcatter Links Under Mobility, Frequency Selectivity, and Antenna Pointing Errors” which was presented at the 2018 IEEE MILCOM conference. He was a recipient of the Purkayastha/TimeLine Ventures graduate fellowship (2002–2003), a co-recipient of the best paper award from the IEEE WPMC 2009 conference, and was a guest editor for the special issue Wireless Cooperative Networks of the EURASIP Journal on Advanced Signal Processing. He was nominated for the MIT TR-35. He has published more than 60 papers in international journals and conferences, made numerous contributions to commercial wireless standards, and is a co-inventor of 10 U.S. patents (granted).

What is your research focus?

Connectivity and security challenges in massive internet-of-things (IoT) devices; Distributed data fusion using cooperative sensing; Autonomy and accountability of cyber-physical systems; Interference management in heterogeneous wireless networks; Coordinated transmission technologies for cellular infrastructure systems; Data-driven and analytics-based self-optimizing wireless networks; Ultra-reliable and low-latency networking for mission-critical systems, and Massive multi-antenna systems leveraging the millimeter-wave unlicensed frequency bands

What is your educational background?

I was trained in the field of Electrical and Computer Engineering with a focus on Communications Theory and Systems.

Recent Publications

Distributed co-phasing with autonomous constellation selection

R. Chopra, R. Annavajjala and C. R. Murthy, “Distributed co-phasing with autonomous constellation selection,” IEEE Trans. Signal Processing, vol. 65, no. 21, pp. 5798-5811, Nov. 2017.

Abstract

In this paper, we consider a distributed cophasing (DCP) technique in which multiple sensor nodes (SNs) communicating their observations to a fusion center (FC) need to transmit different classes of data requiring different levels of error protection. To achieve this, we propose a variant of the DCP technique with autonomous constellation selection at the different SNs. In the first stage of the two-stage time division duplexed cophasing scheme, the SNs obtain estimates of the channels to the FC using pilot symbols transmitted by the latter. Following this, the SNs simultaneously transmit their data symbols prerotated according to the estimated channel phases to combine coherently at the FC. The symbols transmitted by different SNs are drawn from possibly different constellations selected based on the estimated instantaneous channel gains. We show that this scheme is equivalent to transmitting symbols from hierarchical constellations. Based on the properties of hierarchical constellations, we develop recursive expressions for the BER of the proposed system. Following this, we use the properties of the effective channel coefficients to show that it is possible to recover the transmitted data bits from the signal received at the FC blindly, without requiring explicit pilot symbols to be sent by the power-starved SNs. We develop three blind channel estimation and data detection schemes for the presented system model. Using Monte Carlo simulations, we show that the proposed blind channel estimation algorithm achieves a probability of error performance close to that with genie aided perfect channel state information (CSI) at the FC, while using only a moderate number of unknown data symbols for channel estimation.

Multi-stream distributed co-phasing

R. Chopra, C. R. Murthy and R. Annavajjala, “Multi-stream distributed co-phasing,” IEEE Trans. Signal Processing, vol. 65, no. 4, pp. 1042-1057, Feb. 2017.

Abstract:

In this paper, we develop a distributed cophasing (DCP) technique for physical layer fusion of multiple data streams in a wireless sensor network with multiple destination nodes (DNs). The DNs can either be connected to a fusion center (referred to as centralized data processing; CDP) or process data independently and communicate with each other via a rate-limited link (referred to as distributed data processing; DDP). In the first stage of this two-stage cophasing scheme, sensors estimate the channel to the DNs using pilot symbols transmitted by the latter; following which they simultaneously transmit multiple streams of data symbols by prerotating them according to the estimated channel phases to the different DNs. The achievable rates for both CDP and DDP are derived to quantify the gains obtainable by the multistream DCP. In order to aid data detection at the receiver, we propose a least-squares-based iterative algorithm for blind channel estimation in CDP-DCP. Following this, we develop a message passing based blind channel estimation algorithm for DDP-DCP. It is found using Monte Carlo simulations that for the CDP system, the proposed blind channel estimation algorithm achieves a probability of error performance very close to that with perfect CSI at the DNs, while using only a moderate number of unknown data symbols for channel estimation. We also derive approximate expressions for the error probability performance of the proposed system for both CDP and DDP and validate their accuracy using Monte Carlo simulations.

Previous Publications

2017

Integrated acquisition and tracking scheme for channel estimation in millimeter wave wireless networks

L. S. Pillutla and R. Annavajjala, "Integrated acquisition and tracking scheme for channel estimation in millimeter wave wireless networks," in Proc. IEEE 28Th Annual Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), Oct. 2017.

Abstract

In this paper we consider performance of an integrated beam acquisition and tracking scheme for channel estimation in the millimeter-wave (mmWave) wireless networks. By assuming a block based transmission scheme in which pilot symbols are interspersed along with data slots we propose a two stage acquisition scheme namely successive interference cancellation plus iterative regularized least squares (SIC+IRLS) scheme that can be used for acquiring/estimating angle-of-arrival (AoA), angle-of-departure (AoD) and multi-path gains at the pilot slots. To facilitate channel tracking in data slots we also propose a particle filter (PF) based algorithm. Our simulation results demonstrate the superior performance of the integrated (SIC+IRLS)-PF approach as against that of the (SIC+IRLS)-Kalman filter (KF) approach, especially in fast varying conditions.

Bayesian CRLB for Joint AoA, AoD and Multipath Gain Estimation in Millimeter Wave Wireless Networks

L. S. Pillutla and R. Annavajjala, "Bayesian CRLB for Joint AoA, AoD and Multipath Gain Estimation in Millimeter Wave Wireless Networks," in Proc. IEEE Global Communications Conference (GLOBECOM), Dec. 2017.

Abstract

In this paper, we present an analysis of the nonrandom and the Bayesian Cramer-Rao lower bound (CRLB) for the joint estimation of angle-of-arrival (AoA), angle-of-departure (AoD), and the multipath gain in the millimeterwave (mmWave) wireless networks. Our analysis is applicable to multipath channels with Gaussian noise and independent path parameters. Numerical results based on uniform AoA and AoD in [0, π), and Rician fading path gains, reveal that the Bayesian CRLB decreases monotonically with an increase in the Rice factor. Further, the CRLB obtained by using beam forming and combining code books generated by quantizing directly the domain of AoA and AoD was found to be lower than those obtained with other types of beam forming and combining code books.

2016

Multi-stream distributed co-phasing: Design and analysis

R. Chopra, C. R. Murthy, and R. Annavajjala, “Multi-stream distributed co-phasing: Design and analysis,’’ in Proc. IEEE Signal Processing Advances in Wireless Communications (SPAWC), July 2016.

Abstract:

In this paper, we develop and analyze a distributed co-phasing technique for physical layer fusion of common data across a wireless sensor network using multi stream communication, when the fusion center (FC) is equipped with multiple receive antennas. In this two-phase scheme, sensors first estimate their respective channels to the FC antennas by using pilots transmitted by the FC; following which they simultaneously transmit multiple streams of data symbols by pre-rotating them according to the estimated channel phases to the different receive antennas. The mutual information between transmit nodes and the FC is presented, which helps to quantify the gains obtainable by multi stream communication. In order to aid data detection at the receiver, we propose an iterative blind channel estimation algorithm at the FC. It is found using simulations that the proposed blind channel estimation algorithm achieves a symbol error probability (SEP) close to that with perfect channel state information (CSI) at the FC. The SEP of the proposed scheme is theoretically analyzed. The system performance is evaluated via numerical simulation and is observed to be in close agreement with the derived expressions.

Low-complexity distributed algorithms for uplink CoMP in heterogeneous LTE networks

R. Annavajjala, “Low-complexity distributed algorithms for uplink CoMP in heterogeneous LTE networks,” in the special issue on Cloud Radio Access Networks, Journal on Communication and Networks, Apr. 2016.

Abstract

Coordinated multi-point transmission (CoMP) techniques are being touted as enabling technologies for interference mitigation in next generation heterogeneous wireless networks (HetNets). In this paper, we present a comparative performance study of uplink (UL) CoMP algorithms for the 3GPP LTE HetNets. Focusing on a distributed and functionally-split architecture, we consider six distinct UL-CoMP algorithms: 1. Joint reception in the frequency-domain (JRFD) 2. Two-stage equalization (TSEQ) 3. Log-likelihood ratio exchange (LLR-E) 4. Symmetric TSEQ (S-TSEQ) 5. Transport block selection diversity (TBSD) 6. Coordinated scheduling with adaptive interference mitigation (CS-AIM) where JRFD, TSEQ, S-TSEQ, TBSD and CS-AIM are our main contributions in this paper, and quantify their relative performances via the post-processing signal-to-interference-plus-noise ratio distributions. We also compare the CoMP-specific front-haul rate requirements for all the schemes considered in this paper. Our results indicate that, with a linear minimum mean-square error receiver, the JRFD and TSEQ have identical performances, whereas S-TSEQ relaxes the front-haul latency requirements while approaching the performance of TSEQ. Furthermore, in a HetNet environment, we find that CS-AIM provides an attractive alternative to TBSD and LLR-E with a significantly reduced CoMP-specific front-haul rate requirement.

2015

Communication over non-Gaussian channels – Part I: Mutual information and optimum signal detection

R. Annavajjala, C. Yu and J. Zagami, “Communication over non-Gaussian channels—Part I: Information rates and optimum signal detection,” in Proc. IEEE MILCOM 2015, October, 2015.

Abstract:

Gaussian distribution as a source of additive noise is a wellknown modeling assumption, which lead to important and insightful results on the channel capacity, detection reliability, and estimation accuracy. It is also well-recognized that in many practical settings the Gaussian distribution may not be adequate to model the underlying noise phenomenon, and various non-Gaussian noise models are employed for the design and performance analysis of communication systems. However, existing non-Gaussian noise models can be cumbersome to use as it is difficult to prescribe the model parameters and estimate them prior to signal detection, and hard to perform analysis and optimization of communication systems with realistic channel estimation. With the above challenges in mind, in the first part of our work, we introduce a model for the additive non-Gaussian channel (ANGC) with a focus on simplicity, robustness, and ease of analytical tractability. Specifically, we consider a compound noise distribution that is composed of a conditionally additive white Gaussian noise (AWGN) and a Gamma distributed noise variance. This model reduces to the traditional AWGN model when the shape parameter of the Gamma distribution tends to infinity, and the detection and estimation performances on ANGC can easily be reduced to their counterparts on the AWGN channel. With this model, we first present the average mutual information when multiple antennas are employed at the transmitter and the receiver. Next, we derive closed-form expressions for the average probability of error for binary and higher-order modulations when the receiver has perfect channel state information. In the second part of our work, we consider noise variance estimation, pilot-assisted channel estimation (PACE), signal detection with PACE, and coded sequence detection with a mismatched decoding metric.

Communication over non-Gaussian channels – Part II: Channel estimation, mismatched receivers, and error performance with coding

R. Annavajjala, C. Yu and J. Zagami, “Communication over non-Gaussian channels—Part II: Channel estimation, mismatched receivers and error performance with coding,” Proc. IEEE MILCOM 2015, October 2015.

Abstract:

Limited wireless spectrum coupled with the need to support co-existence mechanisms call for robust receivers and improved interference management. With a spectrum sharing operating mode, it is hard to model dynamic interference conditions in wireless networks. In the first part of this work, we have introduced an additive non-Gaussian channel (ANGC) model that captures a variety of interference sources, and presented average mutual information and average probability of error expressions when the receiver has perfect channel estimates and no knowledge of the noise statistics. The primary goal of introducing this model is to compactly represent a wide range of non-Gaussian noise sources in a robust and simple manner. As a continuation of that work, in this paper, we consider the problem of noise variance estimation, signal detection with pilot-assisted channel estimation (PACE), and coded sequence detection with a mismatched decoding metric. We derive the joint maximum-likelihood estimates of the Gaussian noise variance and the non-Gaussian parameter, and expressions for the outage probability and the average bit error probability for uncoded binary signaling. We also derive expressions for the average pairwise error probability with coding and PACE on ANGC. Numerical and simulation results are presented to validate the analysis presented in this paper.

An online learning approach to throughput optimization in wireless networks under dynamic and unknown interference conditions

R. Annavajjala, C. Yu and J. Zagami, “An online learning approach to throughput optimization in wireless networks over dynamic and unknown interference conditions,” in Proc. IEEE Machine Learning in Signal Processing (MLSP) Conference, September 2015.

Abstract:

In this paper, we consider a multi-user communication system with dynamically varying interference on block fading channels. We focus on a multi-antenna receiver, single-antenna transmitters, and the case in which the receiver has no knowledge of the channel state information, interference dynamics, and the variance of the additive noise. Pilot-assisted transmission techniques are employed to enable channel estimation at the receiver. For a given channel coherence length, increasing the number of pilots improves the estimation accuracy, with the tradeoff of reduction in data throughput. Thus, we propose to optimize the pilot content within the data frame to maximize the average data throughput. We employ well-known cross-validation techniques from the machine learning literature to simultaneously improve the estimation accuracy as well as the average throughput. Simulation results with the proposed approach suggest that even when the average number of active interferers is larger than the number of degrees of freedom, at least 85% of the ideal throughput can be achieved with the optimum pilot overhead.

Physical Layer Data Fusion Via Distributed Co-Phasing With General Signal Constellations

A. Manesh, C. R. Murthy, and R. Annavajjala, “Physical layer data fusion via distributed co phasing with general signal constellations,” IEEE Trans. Signal Processing, vol. 63, no. 17, pp. 4660-4672, September 2015.

Abstract:

This paper studies a pilot-assisted physical layer data fusion technique known as Distributed Co-Phasing (DCP). In this two-phase scheme, the sensors first estimate the channel to the fusion center (FC) using pilots sent by the latter; and then they simultaneously transmit their common data by pre-rotating them by the estimated channel phase, thereby achieving physical layer data fusion. First, by analyzing the symmetric mutual information of the system, it is shown that the use of higher order constellations (HOC) can improve the throughput of DCP compared to the binary signaling considered heretofore. Using an HOC in the DCP setting requires the estimation of the composite DCP channel at the FC for data decoding. To this end, two blind algorithms are proposed: (1) power method, and (2) modified K-means algorithm. The latter algorithm is shown to be computationally efficient and converges significantly faster than the conventional K-means algorithm. Analytical expressions for the probability of error are derived, and it is found that even at moderate to low SNRs, the modified K-means algorithm achieves a probability of error comparable to that achievable with a perfect channel estimate at the FC, while requiring no pilot symbols to be transmitted from the sensor nodes. Also, the problem of signal corruption due to imperfect DCP is investigated, and constellation shaping to minimize the probability of signal corruption is proposed and analyzed. The analysis is validated, and the promising performance of DCP for energy-efficient physical layer data fusion is illustrated, using Monte Carlo simulations.

Analysis of Error Probability with Maximum Likelihood Detection over Discrete-Time Memoryless Noncoherent Rayleigh Fading Channels

R. Annavajjala and C. R. Murthy, “Analysis of error probability with maximum likelihood detection over discrete-time memoryless noncoherent Rayleigh fading channels,” in Proc. IEEE Vehicular Technology Conference (VTC-Fall), September 2015.

Abstract

It is known that the capacity of the discrete-time memoryless noncoherent Rayleigh fading channels (DTM-NRFC) is achieved by a discrete constellation with finite number of mass points and when one of the mass points is located at the origin [1]. In this paper, we present the maximum likelihood detection (MLD) error performance on DTM-NRFC for a discrete constellation with coding and spatial diversity. In the absence of outer coding, the error probability with MLD is derived in a surprisingly simple closed-form. On the other hand, with coding and diversity, our error probability expressions can be evaluated via saddle-point approximation techniques.