Title

Dynamic Memory Management For Embedded Non-Volatile Memory

Partners

eVaderis (Grenoble Fr) and Verimag laboratory (Grenoble Fr)

You will mostly be located in the CITI Lab in Lyon but you will have to travel to Grenoble several times during the project (expenses covered)

Funding

3-years PhD grant «Allocation Doctorale de Recherche» from Région Rhône-Alpes

Start date

October 2016

Contact

Guillaume Salagnac

guillaume.salagnac@insa-lyon.fr

More info

perso.citi.insa-lyon.fr/gsalagnac/files/phd-nvram.pdf

Short summary

The general context of this project is smart objects, aka the Internet of Things. Technologically speaking, we are interested in next generation System-on-Chip platforms, based on non-volatile memory technologies. Such technologies make it possible to design very energy-efficient embedded systems, but they also require significant changes in terms of software programming. We aim at designing and studying novel memory management mechanisms, both hardware and software, to improve the performance of such systems. More precisely, we will address the problem of dynamic memory management, i.e. allocating, placing, and moving around the various data structures used by the application program.

Recruitment process

Your profile

Master’s degree or equivalent in the area of computer science / computer engineering

Required skills

Software: Low-level and/or embedded programming in C and/or assembly

Hardware: system-on-chip architecture: processor, memory hierarchy

Languages: fluent English, both written and spoken ; French is a plus but is not required

Bonus skills

Software: discrete event simulation

Work environment: Linux with command-line, scripting language(s)

Personal: autonomy, sense of organization, curiosity, initiative

Relational: good communication skills, intellectual rigour

How to apply

Send an e-mail with one PDF attachment, containing:

Detailed curriculum vitae

Application letter and/or your last internship report

Academic transcripts (including ranking if available) for the last two years of study

2 to 3 letters of recommendation, or a list of reference persons and their e-mail addresses

Title

Non-asymptotic fundamental limits of impulsive radio communications

Laboratory

CITI Lab. Centre of Innovation in Telecommunications and Integration of Service (www.citilab.fr)

Funding

This PhD position takes part of the ANR project ARBurst that has been submitted to the ANR2015. The position will be fully funded if the project is accepted. The final answer for the funding will be known in July 2015.

Contacts

Prof. Jean-Marie Gorce

jean-marie.gorce@insa-lyon.fr

perso.citi.insa-lyon.fr/jmgorce

Dr. Philippe Mary

philippe.mary@insa-rennes.fr

pmary.perso.insa-rennes.fr

Topic

With the recent development of machine-to-machine (M2M) communications and internet-of-things (IoT) networks, the infrastructures have to support more users (or nodes) but each of them requesting a very small quantity of information. This project aims at defining a more appropriate formalism allowing to estimate the theoretical limits of M2M communications. The performance of large scale networks has been widely studied during the past 10 years with usual theoretical tools such as Shannon theory or stochastic geometry. These tools provided interesting insights about scaling laws and theoretical limits but with a limited applicability in the context of M2M, IoT and future 5G services due to the inherent spurious and bursty nature of the associated information flows. While the small packet size invalidates the use of the asymptotic Shannon capacity as a performance indicator, the consequent bursty nature also invalidates the Gaussian assumption usually used to model the interference distribution. As a consequence fundamental limits are neither well known nor even well formulated. The goal of the PhD is to propose new design criteria for M2M networks based on the non-asymptotic information theory framework [1] but taking into account bursty communications, i.e. use of non-Gaussian interference distribution [2], and large-scale deployment, i.e. use of stochastic geometry tool [3]. The candidate will first address the problem of the non-asymptotic bounds (achievability and converse) in a non-Gaussian peer-to-peer (P2P) link. The impulsive noise could be represented by an alpha-stable distribution or other distribution able to capture the impulsiveness of the noise. The Polyanskiy’s approach will be investigated trough the κβ bound method for the achievability part. One the challenge would be to derive an expression (or compute) the dispersion of the impulsive channel [1]. The MolavianJazi’s method [4], based on the central limit theorem (CLT) for functions, could also be investigated in order to approach the mutual information density for a stable noise. The inherent dependence between the rate and the error-probability in finite blocklength regime will help us to define a multiobjective framework for the evaluation of the M2M network performances. Based on these results, the PhD candidate will extend the previous approach to the multi-user case, through the study of the multiple access channel (MAC) and broadcast channel (BC). Based on the outage-splitting theorem for Gaussian MAC [4], the candidate will address the problem of the achievable region of MAC in impulsive noise. The BC scenario will be investigated as a next step. Generally, i.e. in Gaussian framework, the achievable region of multi-user communications is derived under finite second-order moment. This assumption does not hold generally in impulsive noise, overall if alpha-stable distributions are considered, alternative constraint-cost functions need to be considered. A part of the research will consist to clearly define on which assumptions the achievability can be studied in bursty M2M context. Finally, interfering users will be considered as (non-gaussian) noise, but distributed as a Poisson point process (PPP). The challenge is to merge the non-asymptotic theory to the stochastic geometry tool in order to figure out how the channel dispersion behaves in a randomly deployed network and when the interference is considered as non-Gaussian. The work proposed in this PhD could be of a great importance for industrial actors and researchers in the deployment of the future IoT networks. The limits derived in the thesis could provide guidelines to sustain the dramatic increase of the number of connected devices by giving a set of design criteria for these networks.

Key skills

The candidate should have earned an MSc degree, or equivalent, in one of the following field: information theory, signal processing, electrical engineering, applied mathematics. He should have a strong background in probabilities and information theory as well as in signal processing for wireless communications. The candidate should be familiar with Matlab and C/C++ languages.

Key words

Asymptotic and non-asymptotic information theory, capacity, second-order rate, probabilities, mutual information, measure theory, Poisson point process, alpha-stable.

How to apply

– Email a motivation letter
– Full CV with project and courses that could be related to the subject
– Complete academic records (from Bachelor to MSc)
– 2 or 3 references
– Deadline for application September 15th, 2016

References

[1] Y. Polyanskiy, H. V. Poor and S. Verdu, “Channel coding rate in the finite blocklength regime”, IEEE Transactions on Information Theory, vol. 56, no. 5, pp. 2307-2359, May 2010.
[2] G. Samorodnitsky and M. S. Taqqu, Stable Non-Gaussian Random Processes: Stochastic Models with Infinite Variance, Chapmann and Hall, 1994.
[3] F. Baccelli and B. Blaszczyszyn, “Stochastic geometry and wireless networks: volume 1 theory”, Foundations and Trends in Networking, Vol. 3, No. 3-4, pp. 249-449, 2010.
[4] E. MolavianJazi and J. N. Laneman. “A finite blocklength perspective on Gaussian multi-access channels”, CoRR, abs/1309.2343, 2013.

Title

Distributed coding for dense IoT networks

Laboratory

CITI Lab. Centre of Innovation in Telecommunications and Integration of Service (www.citilab.fr)

Funding

This PhD will be supported by the chair Spie-ICS – INSA on IoT that will start in September 2016.

Contact

Prof. Jean-Marie Gorce

jean-marie.gorce@insa-lyon.fr

perso.citi.insa-lyon.fr/jmgorce

Topic

TWith the recent development of machine-to-machine (M2M) communications and internet-of-things (IoT) networks, the cellular network paradigm is evolving to serve massive IoT with bursty traffic in either uplink or downlink mode [3]. Although different techniques are already proposed on the market [2], the best strategy is not known. Considering this massive IoT network paradigm, classical medium access protocols are not efficient because they require lot of signaling for traffic scheduling and channel state information [4]. Multiplexing techniques such as FTDMA, CDMA or OFDMA would require complex synchronization and handshake mechanisms while random access techniques such as CSMA/CA introduce large overheads. In order to maximize the network life-time while ensuring low latency and high reliability, new distributed transmission techniques are necessary. The challenge is to approach the theoretical limits established for the asymptotic regime in [1]. An ideal technique in uplink mode would let the nodes to transmit randomly at their own convenience without any coordination mechanism but with perfect reliability. In downlink, the BS should be able to transmit simultaneously to several nodes while each node would be able to easily decode its own information without complex processing. To approach these limits, the main part of the computational complexity needs to be put at the base station side (in both downlink and uplink cases) using superposition coding or dirty paper coding in downlink mode and multi-user detection techniques in uplink. Additionally, simple but efficient modulation and coding schemes are necessary [4,5]. Another way may rely on exploiting recent results in distributed estimation theory that provided bounds on the reconstruction errors [6]. In addition, opportunistic cooperative mechanisms and compressive sensing techniques when the sources are correlated are possible candidates to optimize the transmission techniques. In this PhD, the candidate will contribute to the design of a new transmission technique devoted to very small packets (as small as few bits) and will study the fundamental tradeoffs: energy efficiency vs spectral efficiency and reliability vs latency. He will design new PHY/MAC strategies and will derive the theoretical proofs of the system performance. The work proposed in this PhD could be of a great importance for the deployment of future IoT networks.

Key skills

The candidate should have earned an MSc degree, or equivalent, in one of the following field: information theory, signal processing, electrical engineering, applied mathematics. He should have a strong background in probabilities and information theory as well as in signal processing for wireless communications. The candidate should be familiar with Matlab and C/C++ languages.

Key words

Modulation and coding, distributed networks, rateless coding, estimation theory, information theory.

How to apply

– Email a motivation letter
– Full CV with project and courses that could be related to the subject
– Complete academic records (from Bachelor to MSc)
– 2 or 3 references
– Applications will be reviewed when they arrive until one candidate is selected

References

[1] Gorce, JM, Poor H.V. & Kelif, JM, “Spatial Continuum Extensions of Asymmetric Gaussian Channels (Multiple Access and Broadcast)”, https://hal.inria.fr/hal-01265184.
[2] Goursaud, C., Gorce, J. M. (2015). Dedicated networks for IoT: PHY/MAC state of the art and challenges. EAI endorsed transactions on Internet of Things.
[3] Dhillon, H. S., Huang, H., Viswanathan, H., & Valenzuela, R. A. (2014). Fundamentals of throughput maximization with random arrivals for M2M communications. Communications, IEEE Transactions on, 62(11), 4094-4109.
[4] Shirvanimoghaddam, M., Li, Y., Dohler, M., Vucetic, B., & Feng, S. (2015). Probabilistic rateless multiple access for machine-to-machine communication. Wireless Communications, IEEE Transactions on, 14(12), 6815-6826.
[5] Shirvanimoghaddam, M., Dohler, M., & Johnson, S. J. (2016). Massive Multiple Access Based on Superposition Raptor Codes for M2M Communications. arXiv preprint arXiv:1602.05671.
[6] Unsal, A., & Knopp, R. (2015). Distributed Sensing and Transmission of Sporadic Random Samples Over a Multiple-Access Channel. Communications, IEEE Transactions on, 63(10), 3813-3828.