Ba Backscatter kscatter-Ai Aided ded Hy Hybrid id Da Data ta - - PowerPoint PPT Presentation

Ba Backscatter kscatter-Ai Aided ded Hy Hybrid id Da Data ta Offl Offloa

• adin

ding g fo for Wi Wirel eless ess Po Powe wered ed Ed Edge ge Se Sens nsor

• r Ne

Netw tworks

• rks

Yuze Zou, PhD School of Electronic Information and Communications Huazhong University of Science and Technology

IEEE Global Communications Conference

9-13 December 2019 Waikoloa, HI, USA Revolutionizing Communications

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

 Introduction  System Model  Energy Minimization for Hybrid Data

Offloading

 Numerical Results

2/18

Ou Outline tline

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

 Introduction  System Model  Energy Minimization for Hybrid Data

Offloading

 Numerical Results

3/18

Ou Outline tline

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

In Introduct troduction ion

4/18

Img src: https://www.newgenapps.com/blog/influence-of-iot-on-wearable-technology

MEC Server

Cloud Server

Edge Devices

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks



Communication models in the edge sensor networks

 Active Transmission

Relative high transmit rate High energy consumption

 Passive Transmission

Relative low transmit rate Negligible energy consumption

In Introduct troduction ion

5/18

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

 Introduction  System Model  Energy Minimization for Hybrid Data

Offloading

 Numerical Results

6/18

Ou Outline tline

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

Sys System tem Model Model

7/18

𝑇𝑗 𝑇

𝑘

⋯ ⋯ ⋯ 𝑇𝑂 𝑇1

𝑕𝑗 ℎ𝑂 HAP

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

 Time Allocation

 Each sensor is allocated with fixed slot, say 𝑈/𝑂  Each sensor operate in passive and then active mode  Each sensor harvests energy when others in passive mode

8/18

Sys System tem Model Model

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

 Signal Model

 Active communications

𝑠

𝑏,𝑗 = 𝐶 log2 1 + 𝑞𝑏,𝑗 ℎ𝑗 2

𝜏2

 Passive communication

𝑠

𝑞,𝑗 = 𝑠 𝑞 NOTE: 𝑠

𝑏,𝑗 > 𝑠 𝑞,𝑗 holds true in general.

9/18

Sys System tem Model Model

Constant

𝑞𝑏,𝑗 = 𝛾 𝑠

𝑏,𝑗

≜ 2𝑠𝑏,𝑗/𝐶 − 1 𝜏2/ ℎ𝑗 2 ⟺

Power Consumption

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

 Introduction  System Model  Energy Minimization for Hybrid Data

Offloading

 Numerical Results

10/18

Ou Outline tline

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

11/18

En Ener ergy gy Mi Mini nimization mization fo for r Hy Hybr brid id Da Data ta Offload Offloading ing min

𝒖𝒃,𝒖𝒒

෍

𝒋∈𝒪

𝑞0𝑢𝑞,𝑗

• s. t.

𝑢𝑏,𝑗 + 𝑢𝑞,𝑗 ≤ 𝑈/𝑂 𝑚𝑏,𝑗 + 𝑚𝑞,𝑗 ≥ 𝑀𝑗 𝑢𝑏,𝑗 ෨ 𝛾 𝑚𝑏,𝑗 𝑢𝑏,𝑗 ≤ 𝜃 ෍

𝑘∈𝒪𝑗

𝑞0𝑕𝑗

2𝑢𝑞,𝑘

𝑢𝑏,𝑗 ≥ 0, 𝑢𝑞,𝑗 ≥ 0, 𝑚𝑏,𝑗 ≥ 0

Slot limitation Task fulfillment Energy constraint Energy minimization at HAP Physical constraint

෨ 𝛾 𝑠

𝑏,𝑗 = 𝛾 𝑠 𝑏,𝑗 + 𝑞𝑑,𝑗 denotes the energy consumption of sensor in active mode

𝜃: energy harvesting coefficient, 𝒪

𝑗 = 𝒪\ 𝑗

𝑀𝑗: total bits to offload of sensor 𝑗, 𝑚𝑏,𝑗 and 𝑚𝑐,𝑗: bits offloaded via active and passive mode, respectively

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

En Ener ergy gy Mi Mini nimization mization fo for r Hy Hybr brid id Da Data ta Offload Offloading ing

12/18  Distributed reformulation

min

𝒖𝒃,𝒖𝒒

෍

𝒋∈𝒪

𝑞0𝑢𝑞,𝑗

• s. t.

𝑢𝑏,𝑗 + 𝑢𝑞,𝑗 ≤ 𝑈/𝑂 𝑚𝑏,𝑗 + 𝑚𝑞,𝑗 ≥ 𝑀𝑗 𝑢𝑏,𝑗 ෨ 𝛾 𝑚𝑏,𝑗 𝑢𝑏,𝑗 ≤ 𝜃 ෍

𝑘∈𝒪𝑗

𝑞0𝑕𝑗

2𝑢𝑞,𝑘

𝑢𝑏,𝑗 ≥ 0, 𝑢𝑞,𝑗 ≥ 0, 𝑚𝑏,𝑗 ≥ 0 𝐹𝑗

Sensor 𝑗 is aware of its energy supply, denote by 𝐹𝑗 > This motivates us to optimize sensors’ offloading scheme in a distributed manner.

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

En Ener ergy gy Mi Mini nimization mization fo for r Hy Hybr brid id Da Data ta Offload Offloading ing

13/18

min

𝑢𝑏,𝑢𝑞,𝑠𝑏 𝑢𝑞

• s. t. C1 𝑢𝑏 + 𝑢𝑞 ≤ 𝑢

C2 𝑢𝑏𝑠

𝑏 + 𝑢𝑞𝑠 𝑞 ≥ 𝑀

C3 𝑢𝑏𝛾 𝑠

𝑏 + 𝑞𝑑𝑢𝑏 ≤ 𝐹

Closed-form solution can be

• btained for each sensor, thus

achieves 𝒫 1 complexity

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

 Introduction  System Model  Energy Minimization for Hybrid Data

Offloading

 Numerical Results

14/18

Ou Outline tline

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

Nume Numerical rical Resu Results lts

15/18

HAP’s transmit power: 𝑞𝑢 = 100 mW Frame length: 𝑈 = 10 Sensor workload: 𝑀 = 10 kbits Energy harvesting efficiency: 𝜃 = 0.8 Circuit power: 𝑞𝑑 = 1 μW Passive data rate: 𝑠

𝑞 = 5 kbps

Downlink channel gain: 𝑕 2 = −53 dB Uplink channel gain: ℎ 2 = −60 dB Noise power: 𝜏2 = −70 dBm Bandwidth: 𝐶 = 400 kHz # sensors: 𝑂 = 10

 Parameter Settings

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

Nume Numerical rical Resu Results lts

16/18

The HAP’s energy consumption and transmit time allocation

Can not offload the workload in time

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

Nume Numerical rical Resu Results lts

17/18

Workload and transmit time

• f passive

data

• ffloading

• Y. Zou et al. Backscatter-Aided Hybrid Data Offloading for Wireless Powered Edge Sensor Networks

Qu Ques estion tions & An s & Answ swers ers

Tha Thank nk yo you u !

Q&A Q&A

18/18