

Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
About the wireless channel fading and related delay
Typology: Study Guides, Projects, Research
1 / 2
This page cannot be seen from the preview
Don't miss anything!


Massa Ndong, Hamidou Tembine Learning & Game Theory Laboratory, New York University Abu Dhabi
Abstract
Wireless communications consist of transmission and reception of data (infor- mation) which is piggybacked by a spectrum of frequencies. Such spectrum has become insufficient to satisfy the current demand in wireless communications because of the high number of devices vying to access and use the wireless net- work. By increasing the spectrum, the capacity of wireless networks increases. Mm-wave piggyback frequencies (F=28GHz, 73GHz,... ) have been recently considered for additional spectrum to address the issue along with Multiple Input Multiple Output (MIMO). A MIMO channel is formed when N t anten- nas transmit and N r antennas receive for N t ≥ 2 and N r ≥ 2. The capacity (maximum rate) of a MIMO channel is proportional to min(N t, N r). However, mm-wave with MIMO requires channel state information(CSI) at both trans- mitter and receiver antennas to yield an optimal capacity. Since the current state-of-the-art relies on channel estimation, an inherent error is computed in the evaluation of the capacity because of the channel error estimation. Wireless communications use digital communications. The data are encoded into sym- bols in the baseband (low frequencies) before being sent in the wireless channel in the passband (high frequencies+). A symbol is transmitted every (^) W^1 seconds where W is the bandwidth occupied by the baseband signal. The transmitted signal arrived at the receiver via several paths. Thus, the received signal is a sum of weighted versions of the transmitted signal. The weights are called fad- ing gains or complex taps. Each path is characterized by its delay and complex tap. The delay which differs from one path to another, is the travel time from transmitter to receiver. The multipath propagation occur in a continuous time. At the reception of the delayed version on the lth path, only the paths which fall in the delay bin [ (^) Wl − (^21) W , (^) Wl + (^21) W ] are considered in the computation of the complex tap hl expressed as hl =
i aie 2 jπF τi (^) , where j (^2) = −1, F is the
piggyback frequency and τi is the delay of the lth path. Thus, all paths out of the delay bin cannot be considered in the reception of the signal. If such paths carry the symbols which can be decoded at the reception, then hl has to be estimated with the paths which fall in the bin. Moreover, the delay on each path may cause signals in different paths adding up at the reception to cancel each other. The received signal is expressed as y[m] =
l hlx[m^ −^ l], where^ m is the discrete arrival time and x[m − l] is the received symbol at previous time
Preprint submitted to NYUAD Poster Session November 1, 2017
m − l. hl is denoted as complex channel tap and determines CSI. As there is an inherent uncertainty in resolving the paths, there is inherent CSI uncertainty. Each transmit antenna appends a certain power Pt to each transmitted symbol. The seminal work on nding the optimal capacity relies on an estimated hl or assume its perfect knowledge. Instead of delving into channel estimation, our study proposes to nd the maximum ergodic rate by appending the P (^) t∗ which maximizes the minimum rate over a certain number of channel realizations. The problem is de ned as a minmax optimization. The conventional resolution presents a high computational cost, thus a Bregman algorithm is proposed to reduce the number of iterations.
Keywords: Channel distribution uncertainty, stochastic optimization, robust game